Merge pull request #5 from rg3/master

updating again

.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 *2017.12.31*. 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 Issues with outdated version will be rejected.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 *2018.01.07*. 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 Issues with outdated version will be rejected.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 **2017.12.31**
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2018.01.07**
 
 ### 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
@@ -35,7 +35,7 @@ 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 2017.12.31
+[debug] youtube-dl version 2018.01.07
 [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

@@ -231,3 +231,4 @@ John Dong
 Tatsuyuki Ishi
 Daniel Weber
 Kay Bouché
+Yang Hongbo

ChangeLog

@@ -1,9 +1,28 @@
 version <unreleased>
 
 Extractors
++ [weibo] Add extractor (#15079)
+* [bilibili] fix extraction (#15188) 
+
+
+version 2018.01.07
+
+Core
+* [utils] Fix youtube-dl under PyPy3 on Windows
+* [YoutubeDL] Output python implementation in debug header
+
+Extractors
++ [jwplatform] Add support for multiple embeds (#15192)
+* [mitele] Fix extraction (#15186)
++ [motherless] Add support for groups (#15124)
+* [lynda] Relax URL regular expression (#15185)
+* [soundcloud] Fallback to avatar picture for thumbnail (#12878)
 * [youku] Fix list extraction (#15135)
 * [openload] Fix extraction (#15166)
+* [lynda] Skip invalid subtitles (#15159)
+* [twitch] Pass video id to url_result when extracting playlist (#15139)
 * [rtve.es:alacarta] Fix extraction of some new URLs
+* [acast] Fix extraction (#15147)
 
 
 version 2017.12.31

docs/supportedsites.md

@@ -478,6 +478,7 @@
  - **Moniker**: allmyvideos.net and vidspot.net
  - **Morningstar**: morningstar.com
  - **Motherless**
+ - **MotherlessGroup**
  - **Motorsport**: motorsport.com
  - **MovieClips**
  - **MovieFap**

youtube_dl/extractor/bilibili.py

@@ -102,6 +102,7 @@ class BiliBiliIE(InfoExtractor):
                     video_id, anime_id, compat_urlparse.urljoin(url, '//bangumi.bilibili.com/anime/%s' % anime_id)))
             headers = {
                 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
+                'Referer': url
             }
             headers.update(self.geo_verification_headers())
 
@@ -116,10 +117,15 @@ class BiliBiliIE(InfoExtractor):
         payload = 'appkey=%s&cid=%s&otype=json&quality=2&type=mp4' % (self._APP_KEY, cid)
         sign = hashlib.md5((payload + self._BILIBILI_KEY).encode('utf-8')).hexdigest()
 
+        headers = {
+            'Referer': url
+        }
+        headers.update(self.geo_verification_headers())
+
         video_info = self._download_json(
             'http://interface.bilibili.com/playurl?%s&sign=%s' % (payload, sign),
             video_id, note='Downloading video info page',
-            headers=self.geo_verification_headers())
+            headers=headers)
 
         if 'durl' not in video_info:
             self._report_error(video_info)

youtube_dl/extractor/canalplus.py

@@ -4,59 +4,36 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlparse
 from ..utils import (
-    dict_get,
     # ExtractorError,
     # HEADRequest,
     int_or_none,
     qualities,
-    remove_end,
     unified_strdate,
 )
 
 
 class CanalplusIE(InfoExtractor):
-    IE_DESC = 'canalplus.fr, piwiplus.fr and d8.tv'
+    IE_DESC = 'mycanal.fr and piwiplus.fr'
-    _VALID_URL = r'''(?x)
+    _VALID_URL = r'https?://(?:www\.)?(?P<site>mycanal|piwiplus)\.fr/(?:[^/]+/)*(?P<display_id>[^?/]+)(?:\.html\?.*\bvid=|/p/)(?P<id>\d+)'
-                        https?://
-                            (?:
-                                (?:
-                                    (?:(?:www|m)\.)?canalplus\.fr|
-                                    (?:www\.)?piwiplus\.fr|
-                                    (?:www\.)?d8\.tv|
-                                    (?:www\.)?c8\.fr|
-                                    (?:www\.)?d17\.tv|
-                                    (?:(?:football|www)\.)?cstar\.fr|
-                                    (?:www\.)?itele\.fr
-                                )/(?:(?:[^/]+/)*(?P<display_id>[^/?#&]+))?(?:\?.*\bvid=(?P<vid>\d+))?|
-                                player\.canalplus\.fr/#/(?P<id>\d+)
-                            )
-
-                    '''
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/%s/%s?format=json'
     _SITE_ID_MAP = {
-        'canalplus': 'cplus',
+        'mycanal': 'cplus',
         'piwiplus': 'teletoon',
-        'd8': 'd8',
-        'c8': 'd8',
-        'd17': 'd17',
-        'cstar': 'd17',
-        'itele': 'itele',
     }
 
     # Only works for direct mp4 URLs
     _GEO_COUNTRIES = ['FR']
 
     _TESTS = [{
-        'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1192814',
+        'url': 'https://www.mycanal.fr/d17-emissions/lolywood/p/1397061',
         'info_dict': {
-            'id': '1405510',
+            'id': '1397061',
-            'display_id': 'pid1830-c-zapping',
+            'display_id': 'lolywood',
             'ext': 'mp4',
-            'title': 'Zapping - 02/07/2016',
+            'title': 'Euro 2016 : Je préfère te prévenir - Lolywood - Episode 34',
-            'description': 'Le meilleur de toutes les chaînes, tous les jours',
+            'description': 'md5:7d97039d455cb29cdba0d652a0efaa5e',
-            'upload_date': '20160702',
+            'upload_date': '20160602',
         },
     }, {
         # geo restricted, bypassed
@@ -70,64 +47,12 @@ class CanalplusIE(InfoExtractor):
             'upload_date': '20140724',
         },
         'expected_warnings': ['HTTP Error 403: Forbidden'],
-    }, {
-        # geo restricted, bypassed
-        'url': 'http://www.c8.fr/c8-divertissement/ms-touche-pas-a-mon-poste/pid6318-videos-integrales.html?vid=1443684',
-        'md5': 'bb6f9f343296ab7ebd88c97b660ecf8d',
-        'info_dict': {
-            'id': '1443684',
-            'display_id': 'pid6318-videos-integrales',
-            'ext': 'mp4',
-            'title': 'Guess my iep ! - TPMP - 07/04/2017',
-            'description': 'md5:6f005933f6e06760a9236d9b3b5f17fa',
-            'upload_date': '20170407',
-        },
-        'expected_warnings': ['HTTP Error 403: Forbidden'],
-    }, {
-        'url': 'http://www.itele.fr/chroniques/invite-michael-darmon/rachida-dati-nicolas-sarkozy-est-le-plus-en-phase-avec-les-inquietudes-des-francais-171510',
-        'info_dict': {
-            'id': '1420176',
-            'display_id': 'rachida-dati-nicolas-sarkozy-est-le-plus-en-phase-avec-les-inquietudes-des-francais-171510',
-            'ext': 'mp4',
-            'title': 'L\'invité de Michaël Darmon du 14/10/2016 - ',
-            'description': 'Chaque matin du lundi au vendredi, Michaël Darmon reçoit un invité politique à 8h25.',
-            'upload_date': '20161014',
-        },
-    }, {
-        'url': 'http://football.cstar.fr/cstar-minisite-foot/pid7566-feminines-videos.html?vid=1416769',
-        'info_dict': {
-            'id': '1416769',
-            'display_id': 'pid7566-feminines-videos',
-            'ext': 'mp4',
-            'title': 'France - Albanie : les temps forts de la soirée - 20/09/2016',
-            'description': 'md5:c3f30f2aaac294c1c969b3294de6904e',
-            'upload_date': '20160921',
-        },
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://m.canalplus.fr/?vid=1398231',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.d17.tv/emissions/pid8303-lolywood.html?vid=1397061',
-        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        site, display_id, video_id = re.match(self._VALID_URL, url).groups()
 
-        site_id = self._SITE_ID_MAP[compat_urllib_parse_urlparse(url).netloc.rsplit('.', 2)[-2]]
+        site_id = self._SITE_ID_MAP[site]
-
-        # Beware, some subclasses do not define an id group
-        display_id = remove_end(dict_get(mobj.groupdict(), ('display_id', 'id', 'vid')), '.html')
-
-        webpage = self._download_webpage(url, display_id)
-        video_id = self._search_regex(
-            [r'<canal:player[^>]+?videoId=(["\'])(?P<id>\d+)',
-             r'id=["\']canal_video_player(?P<id>\d+)',
-             r'data-video=["\'](?P<id>\d+)'],
-            webpage, 'video id', default=mobj.group('vid'), group='id')
 
         info_url = self._VIDEO_INFO_TEMPLATE % (site_id, video_id)
         video_data = self._download_json(info_url, video_id, 'Downloading video JSON')
@@ -161,7 +86,7 @@ class CanalplusIE(InfoExtractor):
                     format_url + '?hdcore=2.11.3', video_id, f4m_id=format_id, fatal=False))
             else:
                 formats.append({
-                    # the secret extracted ya function in http://player.canalplus.fr/common/js/canalPlayer.js
+                    # the secret extracted from ya function in http://player.canalplus.fr/common/js/canalPlayer.js
                     'url': format_url + '?secret=pqzerjlsmdkjfoiuerhsdlfknaes',
                     'format_id': format_id,
                     'preference': preference(format_id),

youtube_dl/extractor/extractors.py

@@ -609,7 +609,10 @@ from .mofosex import MofosexIE
 from .mojvideo import MojvideoIE
 from .moniker import MonikerIE
 from .morningstar import MorningstarIE
-from .motherless import MotherlessIE
+from .motherless import (
+    MotherlessIE,
+    MotherlessGroupIE
+)
 from .motorsport import MotorsportIE
 from .movieclips import MovieClipsIE
 from .moviezine import MoviezineIE
@@ -1295,6 +1298,10 @@ from .webofstories import (
     WebOfStoriesIE,
     WebOfStoriesPlaylistIE,
 )
+from .weibo import (
+    WeiboIE, 
+    WeiboMobileIE
+)
 from .weiqitv import WeiqiTVIE
 from .wimp import WimpIE
 from .wistia import WistiaIE

youtube_dl/extractor/generic.py

@@ -2708,9 +2708,9 @@ class GenericIE(InfoExtractor):
             return self.url_result(viewlift_url)
 
         # Look for JWPlatform embeds
-        jwplatform_url = JWPlatformIE._extract_url(webpage)
+        jwplatform_urls = JWPlatformIE._extract_urls(webpage)
-        if jwplatform_url:
+        if jwplatform_urls:
-            return self.url_result(jwplatform_url, 'JWPlatform')
+            return self.playlist_from_matches(jwplatform_urls, video_id, video_title, ie=JWPlatformIE.ie_key())
 
         # Look for Digiteka embeds
         digiteka_url = DigitekaIE._extract_url(webpage)

youtube_dl/extractor/jwplatform.py

@@ -23,11 +23,14 @@ class JWPlatformIE(InfoExtractor):
 
     @staticmethod
     def _extract_url(webpage):
-        mobj = re.search(
+        urls = JWPlatformIE._extract_urls(webpage)
-            r'<(?:script|iframe)[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
+        return urls[0] if urls else None
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<(?:script|iframe)[^>]+?src=["\']((?:https?:)?//content\.jwplatform\.com/players/[a-zA-Z0-9]{8})',
             webpage)
-        if mobj:
-            return mobj.group('url')
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/lynda.py

@@ -94,7 +94,15 @@ class LyndaBaseIE(InfoExtractor):
 class LyndaIE(LyndaBaseIE):
     IE_NAME = 'lynda'
     IE_DESC = 'lynda.com videos'
-    _VALID_URL = r'https?://(?:www\.)?(?:lynda\.com|educourse\.ga)/(?:[^/]+/[^/]+/(?P<course_id>\d+)|player/embed)/(?P<id>\d+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?(?:lynda\.com|educourse\.ga)/
+                        (?:
+                            (?:[^/]+/){2,3}(?P<course_id>\d+)|
+                            player/embed
+                        )/
+                        (?P<id>\d+)
+                    '''
 
     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 
@@ -113,6 +121,9 @@ class LyndaIE(LyndaBaseIE):
     }, {
         'url': 'https://educourse.ga/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
         'only_matching': True,
+    }, {
+        'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Willkommen-Grundlagen-guten-Gestaltung/393570/393572-4.html',
+        'only_matching': True,
     }]
 
     def _raise_unavailable(self, video_id):
@@ -257,7 +268,15 @@ class LyndaCourseIE(LyndaBaseIE):
 
     # Course link equals to welcome/introduction video link of same course
     # We will recognize it as course link
-    _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P<coursepath>[^/]+/[^/]+/(?P<courseid>\d+))-\d\.html'
+    _VALID_URL = r'https?://(?:www|m)\.(?:lynda\.com|educourse\.ga)/(?P<coursepath>(?:[^/]+/){2,3}(?P<courseid>\d+))-2\.html'
+
+    _TESTS = [{
+        'url': 'https://www.lynda.com/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.lynda.com/de/Graphic-Design-tutorials/Grundlagen-guten-Gestaltung/393570-2.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

youtube_dl/extractor/mitele.py

@@ -1,13 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import json
 import uuid
 
 from .common import InfoExtractor
 from .ooyala import OoyalaIE
 from ..compat import (
     compat_str,
-    compat_urllib_parse_urlencode,
     compat_urlparse,
 )
 from ..utils import (
@@ -42,31 +42,33 @@ class MiTeleBaseIE(InfoExtractor):
                 duration = int_or_none(mmc.get('duration'))
             for location in mmc['locations']:
                 gat = self._proto_relative_url(location.get('gat'), 'http:')
-                bas = location.get('bas')
+                gcp = location.get('gcp')
-                loc = location.get('loc')
                 ogn = location.get('ogn')
-                if None in (gat, bas, loc, ogn):
+                if None in (gat, gcp, ogn):
                     continue
                 token_data = {
-                    'bas': bas,
+                    'gcp': gcp,
-                    'icd': loc,
                     'ogn': ogn,
-                    'sta': '0',
+                    'sta': 0,
                 }
                 media = self._download_json(
-                    '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
+                    gat, video_id, data=json.dumps(token_data).encode('utf-8'),
-                    video_id, 'Downloading %s JSON' % location['loc'])
+                    headers={
-                file_ = media.get('file')
+                        'Content-Type': 'application/json;charset=utf-8',
-                if not file_:
+                        'Referer': url,
+                    })
+                stream = media.get('stream') or media.get('file')
+                if not stream:
                     continue
-                ext = determine_ext(file_)
+                ext = determine_ext(stream)
                 if ext == 'f4m':
                     formats.extend(self._extract_f4m_formats(
-                        file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
+                        stream + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
                         video_id, f4m_id='hds', fatal=False))
                 elif ext == 'm3u8':
                     formats.extend(self._extract_m3u8_formats(
-                        file_, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                        stream, video_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/motherless.py

@@ -4,8 +4,11 @@ import datetime
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     ExtractorError,
+    InAdvancePagedList,
+    orderedSet,
     str_to_int,
     unified_strdate,
 )
@@ -114,3 +117,86 @@ class MotherlessIE(InfoExtractor):
             'age_limit': age_limit,
             'url': video_url,
         }
+
+
+class MotherlessGroupIE(InfoExtractor):
+    _VALID_URL = 'https?://(?:www\.)?motherless\.com/gv?/(?P<id>[a-z0-9_]+)'
+    _TESTS = [{
+        'url': 'http://motherless.com/g/movie_scenes',
+        'info_dict': {
+            'id': 'movie_scenes',
+            'title': 'Movie Scenes',
+            'description': 'Hot and sexy scenes from "regular" movies... '
+                           'Beautiful actresses fully nude... A looot of '
+                           'skin! :)Enjoy!',
+        },
+        'playlist_mincount': 662,
+    }, {
+        'url': 'http://motherless.com/gv/sex_must_be_funny',
+        'info_dict': {
+            'id': 'sex_must_be_funny',
+            'title': 'Sex must be funny',
+            'description': 'Sex can be funny. Wide smiles,laugh, games, fun of '
+                           'any kind!'
+        },
+        'playlist_mincount': 9,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return (False if MotherlessIE.suitable(url)
+                else super(MotherlessGroupIE, cls).suitable(url))
+
+    def _extract_entries(self, webpage, base):
+        entries = []
+        for mobj in re.finditer(
+                r'href="(?P<href>/[^"]+)"[^>]*>(?:\s*<img[^>]+alt="[^-]+-\s(?P<title>[^"]+)")?',
+                webpage):
+            video_url = compat_urlparse.urljoin(base, mobj.group('href'))
+            if not MotherlessIE.suitable(video_url):
+                continue
+            video_id = MotherlessIE._match_id(video_url)
+            title = mobj.group('title')
+            entries.append(self.url_result(
+                video_url, ie=MotherlessIE.ie_key(), video_id=video_id,
+                video_title=title))
+        # Alternative fallback
+        if not entries:
+            entries = [
+                self.url_result(
+                    compat_urlparse.urljoin(base, '/' + video_id),
+                    ie=MotherlessIE.ie_key(), video_id=video_id)
+                for video_id in orderedSet(re.findall(
+                    r'data-codename=["\']([A-Z0-9]+)', webpage))]
+        return entries
+
+    def _real_extract(self, url):
+        group_id = self._match_id(url)
+        page_url = compat_urlparse.urljoin(url, '/gv/%s' % group_id)
+        webpage = self._download_webpage(page_url, group_id)
+        title = self._search_regex(
+            r'<title>([\w\s]+\w)\s+-', webpage, 'title', fatal=False)
+        description = self._html_search_meta(
+            'description', webpage, fatal=False)
+        page_count = self._int(self._search_regex(
+            r'(\d+)</(?:a|span)><(?:a|span)[^>]+>\s*NEXT',
+            webpage, 'page_count'), 'page_count')
+        PAGE_SIZE = 80
+
+        def _get_page(idx):
+            webpage = self._download_webpage(
+                page_url, group_id, query={'page': idx + 1},
+                note='Downloading page %d/%d' % (idx + 1, page_count)
+            )
+            for entry in self._extract_entries(webpage, url):
+                yield entry
+
+        playlist = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
+
+        return {
+            '_type': 'playlist',
+            'id': group_id,
+            'title': title,
+            'description': description,
+            'entries': playlist
+        }

youtube_dl/extractor/ndr.py

@@ -190,10 +190,12 @@ class NDREmbedBaseIE(InfoExtractor):
             ext = determine_ext(src, None)
             if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
-                    src + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id, f4m_id='hds'))
+                    src + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id,
+                    f4m_id='hds', fatal=False))
             elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    src, video_id, 'mp4', m3u8_id='hls', entry_protocol='m3u8_native'))
+                    src, video_id, 'mp4', m3u8_id='hls',
+                    entry_protocol='m3u8_native', fatal=False))
             else:
                 quality = f.get('quality')
                 ff = {

youtube_dl/extractor/odnoklassniki.py

@@ -19,11 +19,11 @@ from ..utils import (
 
 
 class OdnoklassnikiIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?:www|m|mobile)\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P<id>[\d-]+)'
+    _VALID_URL = r'https?://(?:(?:www|m|mobile)\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer|live)/(?P<id>[\d-]+)'
     _TESTS = [{
         # metadata in JSON
         'url': 'http://ok.ru/video/20079905452',
-        'md5': '6ba728d85d60aa2e6dd37c9e70fdc6bc',
+        'md5': '0b62089b479e06681abaaca9d204f152',
         'info_dict': {
             'id': '20079905452',
             'ext': 'mp4',
@@ -35,7 +35,6 @@ class OdnoklassnikiIE(InfoExtractor):
             'like_count': int,
             'age_limit': 0,
         },
-        'skip': 'Video has been blocked',
     }, {
         # metadataUrl
         'url': 'http://ok.ru/video/63567059965189-0?fromTime=5',
@@ -99,6 +98,9 @@ class OdnoklassnikiIE(InfoExtractor):
     }, {
         'url': 'http://mobile.ok.ru/video/20079905452',
         'only_matching': True,
+    }, {
+        'url': 'https://www.ok.ru/live/484531969818',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -184,6 +186,10 @@ class OdnoklassnikiIE(InfoExtractor):
             })
             return info
 
+        assert title
+        if provider == 'LIVE_TV_APP':
+            info['title'] = self._live_title(title)
+
         quality = qualities(('4', '0', '1', '2', '3', '5'))
 
         formats = [{
@@ -210,6 +216,20 @@ class OdnoklassnikiIE(InfoExtractor):
             if fmt_type:
                 fmt['quality'] = quality(fmt_type)
 
+        # Live formats
+        m3u8_url = metadata.get('hlsMasterPlaylistUrl')
+        if m3u8_url:
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', entry_protocol='m3u8',
+                m3u8_id='hls', fatal=False))
+        rtmp_url = metadata.get('rtmpUrl')
+        if rtmp_url:
+            formats.append({
+                'url': rtmp_url,
+                'format_id': 'rtmp',
+                'ext': 'flv',
+            })
+
         self._sort_formats(formats)
 
         info['formats'] = formats

youtube_dl/extractor/soundcloud.py

@@ -136,6 +136,25 @@ class SoundcloudIE(InfoExtractor):
                 'license': 'all-rights-reserved',
             },
         },
+        # no album art, use avatar pic for thumbnail
+        {
+            'url': 'https://soundcloud.com/garyvee/sideways-prod-mad-real',
+            'md5': '59c7872bc44e5d99b7211891664760c2',
+            'info_dict': {
+                'id': '309699954',
+                'ext': 'mp3',
+                'title': 'Sideways (Prod. Mad Real)',
+                'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+                'uploader': 'garyvee',
+                'upload_date': '20170226',
+                'duration': 207,
+                'thumbnail': r're:https?://.*\.jpg',
+                'license': 'all-rights-reserved',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
     ]
 
     _CLIENT_ID = 'c6CU49JDMapyrQo06UxU9xouB9ZVzqCn'
@@ -160,7 +179,7 @@ class SoundcloudIE(InfoExtractor):
         name = full_title or track_id
         if quiet:
             self.report_extraction(name)
-        thumbnail = info.get('artwork_url')
+        thumbnail = info.get('artwork_url') or info.get('user', {}).get('avatar_url')
         if isinstance(thumbnail, compat_str):
             thumbnail = thumbnail.replace('-large', '-t500x500')
         ext = 'mp3'

youtube_dl/extractor/weibo.py

@@ -0,0 +1,140 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+import json
+import random
+import re
+
+from ..compat import (
+    compat_parse_qs,
+    compat_str,
+)
+from ..utils import (
+    js_to_json,
+    strip_jsonp,
+    urlencode_postdata,
+)
+
+
+class WeiboIE(InfoExtractor):
+    _VALID_URL = r'https?://weibo\.com/[0-9]+/(?P<id>[a-zA-Z0-9]+)'
+    _TEST = {
+        'url': 'https://weibo.com/6275294458/Fp6RGfbff?type=comment',
+        'info_dict': {
+            'id': 'Fp6RGfbff',
+            'ext': 'mp4',
+            'title': 'You should have servants to massage you,... 来自Hosico_猫 - 微博',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        # to get Referer url for genvisitor
+        webpage, urlh = self._download_webpage_handle(url, video_id)
+
+        visitor_url = urlh.geturl()
+
+        if 'passport.weibo.com' in visitor_url:
+            # first visit
+            visitor_data = self._download_json(
+                'https://passport.weibo.com/visitor/genvisitor', video_id,
+                note='Generating first-visit data',
+                transform_source=strip_jsonp,
+                headers={'Referer': visitor_url},
+                data=urlencode_postdata({
+                    'cb': 'gen_callback',
+                    'fp': json.dumps({
+                        'os': '2',
+                        'browser': 'Gecko57,0,0,0',
+                        'fonts': 'undefined',
+                        'screenInfo': '1440*900*24',
+                        'plugins': '',
+                    }),
+                }))
+
+            tid = visitor_data['data']['tid']
+            cnfd = '%03d' % visitor_data['data']['confidence']
+
+            self._download_webpage(
+                'https://passport.weibo.com/visitor/visitor', video_id,
+                note='Running first-visit callback',
+                query={
+                    'a': 'incarnate',
+                    't': tid,
+                    'w': 2,
+                    'c': cnfd,
+                    'cb': 'cross_domain',
+                    'from': 'weibo',
+                    '_rand': random.random(),
+                })
+
+            webpage = self._download_webpage(
+                url, video_id, note='Revisiting webpage')
+
+        title = self._html_search_regex(
+            r'<title>(.+?)</title>', webpage, 'title')
+
+        video_formats = compat_parse_qs(self._search_regex(
+            r'video-sources=\\\"(.+?)\"', webpage, 'video_sources'))
+
+        formats = []
+        supported_resolutions = (480, 720)
+        for res in supported_resolutions:
+            vid_urls = video_formats.get(compat_str(res))
+            if not vid_urls or not isinstance(vid_urls, list):
+                continue
+
+            vid_url = vid_urls[0]
+            formats.append({
+                'url': vid_url,
+                'height': res,
+            })
+
+        self._sort_formats(formats)
+
+        uploader = self._og_search_property(
+            'nick-name', webpage, 'uploader', default=None)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'uploader': uploader,
+            'formats': formats
+        }
+
+
+class WeiboMobileIE(InfoExtractor):
+    _VALID_URL = r'https?://m\.weibo\.cn/status/(?P<id>[0-9]+)(\?.+)?'
+    _TEST = {
+        'url': 'https://m.weibo.cn/status/4189191225395228?wm=3333_2001&sourcetype=weixin&featurecode=newtitle&from=singlemessage&isappinstalled=0',
+        'info_dict': {
+            'id': '4189191225395228',
+            'ext': 'mp4',
+            'title': '午睡当然是要甜甜蜜蜜的啦',
+            'uploader': '柴犬柴犬'
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        # to get Referer url for genvisitor
+        webpage = self._download_webpage(url, video_id, note='visit the page')
+
+        weibo_info = self._parse_json(self._search_regex(
+            r'var\s+\$render_data\s*=\s*\[({.*})\]\[0\]\s*\|\|\s*{};',
+            webpage, 'js_code', flags=re.DOTALL),
+            video_id, transform_source=js_to_json)
+
+        status_data = weibo_info.get('status', {})
+        page_info = status_data.get('page_info')
+        title = status_data['status_title']
+        uploader = status_data.get('user', {}).get('screen_name')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'uploader': uploader,
+            'url': page_info['media_info']['stream_url']
+        }

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2017.12.31'
+__version__ = '2018.01.07'