release 2016.06.23.1

The pattern has changed from "video123412" to "video-o8xa19".
The changes maintain backwards compatibility with old-style URLs.

.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 *2016.06.19.1*. 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 Issues with outdated version will be rejected.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 **2016.06.19.1**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.06.23.1*. 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 Issues with outdated version will be rejected.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 **2016.06.23.1**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2016.06.19.1
+[debug] youtube-dl version 2016.06.23.1
 [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: {}

devscripts/make_lazy_extractors.py

@@ -14,15 +14,17 @@ if os.path.exists(lazy_extractors_filename):
     os.remove(lazy_extractors_filename)
 
 from youtube_dl.extractor import _ALL_CLASSES
-from youtube_dl.extractor.common import InfoExtractor
+from youtube_dl.extractor.common import InfoExtractor, SearchInfoExtractor
 
 with open('devscripts/lazy_load_template.py', 'rt') as f:
     module_template = f.read()
 
-module_contents = [module_template + '\n' + getsource(InfoExtractor.suitable)]
+module_contents = [
+    module_template + '\n' + getsource(InfoExtractor.suitable) + '\n',
+    'class LazyLoadSearchExtractor(LazyLoadExtractor):\n    pass\n']
 
 ie_template = '''
-class {name}(LazyLoadExtractor):
+class {name}({bases}):
     _VALID_URL = {valid_url!r}
     _module = '{module}'
 '''
@@ -34,10 +36,20 @@ make_valid_template = '''
 '''
 
 
+def get_base_name(base):
+    if base is InfoExtractor:
+        return 'LazyLoadExtractor'
+    elif base is SearchInfoExtractor:
+        return 'LazyLoadSearchExtractor'
+    else:
+        return base.__name__
+
+
 def build_lazy_ie(ie, name):
     valid_url = getattr(ie, '_VALID_URL', None)
     s = ie_template.format(
         name=name,
+        bases=', '.join(map(get_base_name, ie.__bases__)),
         valid_url=valid_url,
         module=ie.__module__)
     if ie.suitable.__func__ is not InfoExtractor.suitable.__func__:
@@ -47,12 +59,35 @@ def build_lazy_ie(ie, name):
         s += make_valid_template.format(valid_url=ie._make_valid_url())
     return s
 
+# find the correct sorting and add the required base classes so that sublcasses
+# can be correctly created
+classes = _ALL_CLASSES[:-1]
+ordered_cls = []
+while classes:
+    for c in classes[:]:
+        bases = set(c.__bases__) - set((object, InfoExtractor, SearchInfoExtractor))
+        stop = False
+        for b in bases:
+            if b not in classes and b not in ordered_cls:
+                if b.__name__ == 'GenericIE':
+                    exit()
+                classes.insert(0, b)
+                stop = True
+        if stop:
+            break
+        if all(b in ordered_cls for b in bases):
+            ordered_cls.append(c)
+            classes.remove(c)
+            break
+ordered_cls.append(_ALL_CLASSES[-1])
+
 names = []
-for ie in list(sorted(_ALL_CLASSES[:-1], key=lambda cls: cls.ie_key())) + _ALL_CLASSES[-1:]:
-    name = ie.ie_key() + 'IE'
+for ie in ordered_cls:
+    name = ie.__name__
     src = build_lazy_ie(ie, name)
     module_contents.append(src)
-    names.append(name)
+    if ie in _ALL_CLASSES:
+        names.append(name)
 
 module_contents.append(
     '_ALL_CLASSES = [{0}]'.format(', '.join(names)))

docs/supportedsites.md

@@ -248,7 +248,6 @@
  - **Gamersyde**
  - **GameSpot**
  - **GameStar**
- - **Gametrailers**
  - **Gazeta**
  - **GDCVault**
  - **generic**: Generic downloader that works on some sites
@@ -386,7 +385,6 @@
  - **MovieFap**
  - **Moviezine**
  - **MPORA**
- - **MSNBC**
  - **MTV**
  - **mtv.de**
  - **mtviggy.com**

youtube_dl/downloader/hls.py

@@ -2,14 +2,24 @@ from __future__ import unicode_literals
 
 import os.path
 import re
+import binascii
+try:
+    from Crypto.Cipher import AES
+    can_decrypt_frag = True
+except ImportError:
+    can_decrypt_frag = False
 
 from .fragment import FragmentFD
 from .external import FFmpegFD
 
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_urlparse,
+    compat_struct_pack,
+)
 from ..utils import (
     encodeFilename,
     sanitize_open,
+    parse_m3u8_attributes,
 )
 
 
@@ -21,7 +31,7 @@ class HlsFD(FragmentFD):
     @staticmethod
     def can_download(manifest):
         UNSUPPORTED_FEATURES = (
-            r'#EXT-X-KEY:METHOD=(?!NONE)',  # encrypted streams [1]
+            r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)',  # encrypted streams [1]
             r'#EXT-X-BYTERANGE',  # playlists composed of byte ranges of media files [2]
 
             # Live streams heuristic does not always work (e.g. geo restricted to Germany
@@ -39,7 +49,9 @@ class HlsFD(FragmentFD):
             # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.2
             # 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5
         )
-        return all(not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES)
+        check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES]
+        check_results.append(can_decrypt_frag or '#EXT-X-KEY:METHOD=AES-128' not in manifest)
+        return all(check_results)
 
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
@@ -57,36 +69,60 @@ class HlsFD(FragmentFD):
                 fd.add_progress_hook(ph)
             return fd.real_download(filename, info_dict)
 
-        fragment_urls = []
+        total_frags = 0
         for line in s.splitlines():
             line = line.strip()
             if line and not line.startswith('#'):
-                segment_url = (
-                    line
-                    if re.match(r'^https?://', line)
-                    else compat_urlparse.urljoin(man_url, line))
-                fragment_urls.append(segment_url)
-                # We only download the first fragment during the test
-                if self.params.get('test', False):
-                    break
+                total_frags += 1
 
         ctx = {
             'filename': filename,
-            'total_frags': len(fragment_urls),
+            'total_frags': total_frags,
         }
 
         self._prepare_and_start_frag_download(ctx)
 
+        i = 0
+        media_sequence = 0
+        decrypt_info = {'METHOD': 'NONE'}
         frags_filenames = []
-        for i, frag_url in enumerate(fragment_urls):
-            frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
-            success = ctx['dl'].download(frag_filename, {'url': frag_url})
-            if not success:
-                return False
-            down, frag_sanitized = sanitize_open(frag_filename, 'rb')
-            ctx['dest_stream'].write(down.read())
-            down.close()
-            frags_filenames.append(frag_sanitized)
+        for line in s.splitlines():
+            line = line.strip()
+            if line:
+                if not line.startswith('#'):
+                    frag_url = (
+                        line
+                        if re.match(r'^https?://', line)
+                        else compat_urlparse.urljoin(man_url, line))
+                    frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
+                    success = ctx['dl'].download(frag_filename, {'url': frag_url})
+                    if not success:
+                        return False
+                    down, frag_sanitized = sanitize_open(frag_filename, 'rb')
+                    frag_content = down.read()
+                    down.close()
+                    if decrypt_info['METHOD'] == 'AES-128':
+                        iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence)
+                        frag_content = AES.new(
+                            decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
+                    ctx['dest_stream'].write(frag_content)
+                    frags_filenames.append(frag_sanitized)
+                    # We only download the first fragment during the test
+                    if self.params.get('test', False):
+                        break
+                    i += 1
+                    media_sequence += 1
+                elif line.startswith('#EXT-X-KEY'):
+                    decrypt_info = parse_m3u8_attributes(line[11:])
+                    if decrypt_info['METHOD'] == 'AES-128':
+                        if 'IV' in decrypt_info:
+                            decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:])
+                        if not re.match(r'^https?://', decrypt_info['URI']):
+                            decrypt_info['URI'] = compat_urlparse.urljoin(
+                                man_url, decrypt_info['URI'])
+                        decrypt_info['KEY'] = self.ydl.urlopen(decrypt_info['URI']).read()
+                elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
+                    media_sequence = int(line[22:])
 
         self._finish_frag_download(ctx)
 

youtube_dl/extractor/cbs.py

@@ -1,17 +1,13 @@
 from __future__ import unicode_literals
 
-import re
-
-from .theplatform import ThePlatformIE
+from .theplatform import ThePlatformFeedIE
 from ..utils import (
-    xpath_text,
-    xpath_element,
     int_or_none,
     find_xpath_attr,
 )
 
 
-class CBSBaseIE(ThePlatformIE):
+class CBSBaseIE(ThePlatformFeedIE):
     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
         closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
         return {
@@ -21,9 +17,22 @@ class CBSBaseIE(ThePlatformIE):
             }]
         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
 
+    def _extract_video_info(self, filter_query, video_id):
+        return self._extract_feed_info(
+            'dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id, lambda entry: {
+                'series': entry.get('cbs$SeriesTitle'),
+                'season_number': int_or_none(entry.get('cbs$SeasonNumber')),
+                'episode': entry.get('cbs$EpisodeTitle'),
+                'episode_number': int_or_none(entry.get('cbs$EpisodeNumber')),
+            }, {
+                'StreamPack': {
+                    'manifest': 'm3u',
+                }
+            })
+
 
 class CBSIE(CBSBaseIE):
-    _VALID_URL = r'(?:cbs:(?P<content_id>\w+)|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<display_id>[^/]+))'
+    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
 
     _TESTS = [{
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
@@ -38,25 +47,7 @@ class CBSIE(CBSBaseIE):
             'upload_date': '20131127',
             'uploader': 'CBSI-NEW',
         },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
-        '_skip': 'Blocked outside the US',
-    }, {
-        'url': 'http://www.cbs.com/shows/liveonletterman/artist/221752/st-vincent/',
-        'info_dict': {
-            'id': 'WWF_5KqY3PK1',
-            'display_id': 'st-vincent',
-            'ext': 'flv',
-            'title': 'Live on Letterman - St. Vincent',
-            'description': 'Live On Letterman: St. Vincent in concert from New York\'s Ed Sullivan Theater on Tuesday, July 16, 2014.',
-            'duration': 3221,
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
+        'expected_warnings': ['Failed to download m3u8 information'],
         '_skip': 'Blocked outside the US',
     }, {
         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
@@ -68,44 +59,5 @@ class CBSIE(CBSBaseIE):
     TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
 
     def _real_extract(self, url):
-        content_id, display_id = re.match(self._VALID_URL, url).groups()
-        if not content_id:
-            webpage = self._download_webpage(url, display_id)
-            content_id = self._search_regex(
-                [r"video\.settings\.content_id\s*=\s*'([^']+)';", r"cbsplayer\.contentId\s*=\s*'([^']+)';"],
-                webpage, 'content id')
-        items_data = self._download_xml(
-            'http://can.cbs.com/thunder/player/videoPlayerService.php',
-            content_id, query={'partner': 'cbs', 'contentId': content_id})
-        video_data = xpath_element(items_data, './/item')
-        title = xpath_text(video_data, 'videoTitle', 'title', True)
-
-        subtitles = {}
-        formats = []
-        for item in items_data.findall('.//item'):
-            pid = xpath_text(item, 'pid')
-            if not pid:
-                continue
-            tp_release_url = self.TP_RELEASE_URL_TEMPLATE % pid
-            if '.m3u8' in xpath_text(item, 'contentUrl', default=''):
-                tp_release_url += '&manifest=m3u'
-            tp_formats, tp_subtitles = self._extract_theplatform_smil(
-                tp_release_url, content_id, 'Downloading %s SMIL data' % pid)
-            formats.extend(tp_formats)
-            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
-        self._sort_formats(formats)
-
-        info = self.get_metadata('dJ5BDC/media/guid/2198311517/%s' % content_id, content_id)
-        info.update({
-            'id': content_id,
-            'display_id': display_id,
-            'title': title,
-            'series': xpath_text(video_data, 'seriesTitle'),
-            'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
-            'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
-            'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
-            'thumbnail': xpath_text(video_data, 'previewImageURL'),
-            'formats': formats,
-            'subtitles': subtitles,
-        })
-        return info
+        content_id = self._match_id(url)
+        return self._extract_video_info('byGuid=%s' % content_id, content_id)

youtube_dl/extractor/cbsnews.py

@@ -30,9 +30,12 @@ class CBSNewsIE(CBSBaseIE):
         {
             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
             'info_dict': {
-                'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
+                'id': 'SNJBOYzXiWBOvaLsdzwH8fmtP1SCd91Y',
                 'ext': 'mp4',
                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
+                'description': 'md5:4a6983e480542d8b333a947bfc64ddc7',
+                'upload_date': '19700101',
+                'uploader': 'CBSI-NEW',
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 205,
                 'subtitles': {
@@ -58,30 +61,8 @@ class CBSNewsIE(CBSBaseIE):
             webpage, 'video JSON info'), video_id)
 
         item = video_info['item'] if 'item' in video_info else video_info
-        title = item.get('articleTitle') or item.get('hed')
-        duration = item.get('duration')
-        thumbnail = item.get('mediaImage') or item.get('thumbnail')
-
-        subtitles = {}
-        formats = []
-        for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
-            pid = item.get('media' + format_id)
-            if not pid:
-                continue
-            release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true' % pid
-            tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid)
-            formats.extend(tp_formats)
-            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
+        guid = item['mpxRefId']
+        return self._extract_video_info('byGuid=%s' % guid, guid)
 
 
 class CBSNewsLiveVideoIE(InfoExtractor):

youtube_dl/extractor/cbssports.py

@@ -1,30 +1,28 @@
 from __future__ import unicode_literals
 
-import re
-
-from .common import InfoExtractor
+from .cbs import CBSBaseIE
 
 
-class CBSSportsIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.cbssports\.com/video/player/(?P<section>[^/]+)/(?P<id>[^/]+)'
+class CBSSportsIE(CBSBaseIE):
+    _VALID_URL = r'https?://www\.cbssports\.com/video/player/[^/]+/(?P<id>\d+)'
 
-    _TEST = {
-        'url': 'http://www.cbssports.com/video/player/tennis/318462531970/0/us-open-flashbacks-1990s',
+    _TESTS = [{
+        'url': 'http://www.cbssports.com/video/player/videos/708337219968/0/ben-simmons-the-next-lebron?-not-so-fast',
         'info_dict': {
-            'id': '_d5_GbO8p1sT',
-            'ext': 'flv',
-            'title': 'US Open flashbacks: 1990s',
-            'description': 'Bill Macatee relives the best moments in US Open history from the 1990s.',
+            'id': '708337219968',
+            'ext': 'mp4',
+            'title': 'Ben Simmons the next LeBron? Not so fast',
+            'description': 'md5:854294f627921baba1f4b9a990d87197',
+            'timestamp': 1466293740,
+            'upload_date': '20160618',
+            'uploader': 'CBSI-NEW',
         },
-    }
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        section = mobj.group('section')
-        video_id = mobj.group('id')
-        all_videos = self._download_json(
-            'http://www.cbssports.com/data/video/player/getVideos/%s?as=json' % section,
-            video_id)
-        # The json file contains the info of all the videos in the section
-        video_info = next(v for v in all_videos if v['pcid'] == video_id)
-        return self.url_result('theplatform:%s' % video_info['pid'], 'ThePlatform')
+        video_id = self._match_id(url)
+        return self._extract_video_info('byId=%s' % video_id, video_id)

youtube_dl/extractor/common.py

@@ -53,6 +53,7 @@ from ..utils import (
     mimetype2ext,
     update_Request,
     update_url_query,
+    parse_m3u8_attributes,
 )
 
 
@@ -1150,23 +1151,11 @@ class InfoExtractor(object):
             }]
         last_info = None
         last_media = None
-        kv_rex = re.compile(
-            r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
         for line in m3u8_doc.splitlines():
             if line.startswith('#EXT-X-STREAM-INF:'):
-                last_info = {}
-                for m in kv_rex.finditer(line):
-                    v = m.group('val')
-                    if v.startswith('"'):
-                        v = v[1:-1]
-                    last_info[m.group('key')] = v
+                last_info = parse_m3u8_attributes(line)
             elif line.startswith('#EXT-X-MEDIA:'):
-                last_media = {}
-                for m in kv_rex.finditer(line):
-                    v = m.group('val')
-                    if v.startswith('"'):
-                        v = v[1:-1]
-                    last_media[m.group('key')] = v
+                last_media = parse_m3u8_attributes(line)
             elif line.startswith('#') or not line.strip():
                 continue
             else:

youtube_dl/extractor/extractors.py

@@ -285,7 +285,6 @@ from .gameone import (
 from .gamersyde import GamersydeIE
 from .gamespot import GameSpotIE
 from .gamestar import GameStarIE
-from .gametrailers import GametrailersIE
 from .gazeta import GazetaIE
 from .gdcvault import GDCVaultIE
 from .generic import GenericIE
@@ -481,7 +480,6 @@ from .nbc import (
     NBCNewsIE,
     NBCSportsIE,
     NBCSportsVPlayerIE,
-    MSNBCIE,
 )
 from .ndr import (
     NDRIE,

youtube_dl/extractor/facebook.py

@@ -239,6 +239,8 @@ class FacebookIE(InfoExtractor):
 
         formats = []
         for format_id, f in video_data.items():
+            if f and isinstance(f, dict):
+                f = [f]
             if not f or not isinstance(f, list):
                 continue
             for quality in ('sd', 'hd'):

youtube_dl/extractor/foxsports.py

@@ -1,7 +1,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import smuggle_url
+from ..utils import (
+    smuggle_url,
+    update_url_query,
+)
 
 
 class FoxSportsIE(InfoExtractor):
@@ -9,11 +12,15 @@ class FoxSportsIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.foxsports.com/video?vid=432609859715',
+        'md5': 'b49050e955bebe32c301972e4012ac17',
         'info_dict': {
-            'id': 'gA0bHB3Ladz3',
-            'ext': 'flv',
+            'id': 'i0qKWsk3qJaM',
+            'ext': 'mp4',
             'title': 'Courtney Lee on going up 2-0 in series vs. Blazers',
             'description': 'Courtney Lee talks about Memphis being focused.',
+            'upload_date': '20150423',
+            'timestamp': 1429761109,
+            'uploader': 'NEWA-FNG-FOXSPORTS',
         },
         'add_ie': ['ThePlatform'],
     }
@@ -28,5 +35,8 @@ class FoxSportsIE(InfoExtractor):
                 r"data-player-config='([^']+)'", webpage, 'data player config'),
             video_id)
 
-        return self.url_result(smuggle_url(
-            config['releaseURL'] + '&manifest=f4m', {'force_smil_url': True}))
+        return self.url_result(smuggle_url(update_url_query(
+            config['releaseURL'], {
+                'mbr': 'true',
+                'switch': 'http',
+            }), {'force_smil_url': True}))

youtube_dl/extractor/gamespot.py

@@ -1,19 +1,19 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
-from .common import InfoExtractor
+from .once import OnceIE
 from ..compat import (
     compat_urllib_parse_unquote,
-    compat_urlparse,
 )
 from ..utils import (
     unescapeHTML,
+    url_basename,
+    dict_get,
 )
 
 
-class GameSpotIE(InfoExtractor):
+class GameSpotIE(OnceIE):
     _VALID_URL = r'https?://(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
     _TESTS = [{
         'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
@@ -39,29 +39,73 @@ class GameSpotIE(InfoExtractor):
         webpage = self._download_webpage(url, page_id)
         data_video_json = self._search_regex(
             r'data-video=["\'](.*?)["\']', webpage, 'data video')
-        data_video = json.loads(unescapeHTML(data_video_json))
+        data_video = self._parse_json(unescapeHTML(data_video_json), page_id)
         streams = data_video['videoStreams']
 
+        manifest_url = None
         formats = []
         f4m_url = streams.get('f4m_stream')
-        if f4m_url is not None:
-            # Transform the manifest url to a link to the mp4 files
-            # they are used in mobile devices.
-            f4m_path = compat_urlparse.urlparse(f4m_url).path
-            QUALITIES_RE = r'((,\d+)+,?)'
-            qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
-            http_path = f4m_path[1:].split('/', 1)[1]
-            http_template = re.sub(QUALITIES_RE, r'%s', http_path)
-            http_template = http_template.replace('.csmil/manifest.f4m', '')
-            http_template = compat_urlparse.urljoin(
-                'http://video.gamespotcdn.com/', http_template)
-            for q in qualities:
-                formats.append({
-                    'url': http_template % q,
-                    'ext': 'mp4',
-                    'format_id': q,
-                })
-        else:
+        if f4m_url:
+            manifest_url = f4m_url
+            formats.extend(self._extract_f4m_formats(
+                f4m_url + '?hdcore=3.7.0', page_id, f4m_id='hds', fatal=False))
+        m3u8_url = streams.get('m3u8_stream')
+        if m3u8_url:
+            manifest_url = m3u8_url
+            m3u8_formats = self._extract_m3u8_formats(
+                m3u8_url, page_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False)
+            formats.extend(m3u8_formats)
+        progressive_url = dict_get(
+            streams, ('progressive_hd', 'progressive_high', 'progressive_low'))
+        if progressive_url and manifest_url:
+            qualities_basename = self._search_regex(
+                '/([^/]+)\.csmil/',
+                manifest_url, 'qualities basename', default=None)
+            if qualities_basename:
+                QUALITIES_RE = r'((,\d+)+,?)'
+                qualities = self._search_regex(
+                    QUALITIES_RE, qualities_basename,
+                    'qualities', default=None)
+                if qualities:
+                    qualities = list(map(lambda q: int(q), qualities.strip(',').split(',')))
+                    qualities.sort()
+                    http_template = re.sub(QUALITIES_RE, r'%d', qualities_basename)
+                    http_url_basename = url_basename(progressive_url)
+                    if m3u8_formats:
+                        self._sort_formats(m3u8_formats)
+                        m3u8_formats = list(filter(
+                            lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
+                            m3u8_formats))
+                    if len(qualities) == len(m3u8_formats):
+                        for q, m3u8_format in zip(qualities, m3u8_formats):
+                            f = m3u8_format.copy()
+                            f.update({
+                                'url': progressive_url.replace(
+                                    http_url_basename, http_template % q),
+                                'format_id': f['format_id'].replace('hls', 'http'),
+                                'protocol': 'http',
+                            })
+                            formats.append(f)
+                    else:
+                        for q in qualities:
+                            formats.append({
+                                'url': progressive_url.replace(
+                                    http_url_basename, http_template % q),
+                                'ext': 'mp4',
+                                'format_id': 'http-%d' % q,
+                                'tbr': q,
+                            })
+
+        onceux_json = self._search_regex(
+            r'data-onceux-options=["\'](.*?)["\']', webpage, 'data video', default=None)
+        if onceux_json:
+            onceux_url = self._parse_json(unescapeHTML(onceux_json), page_id).get('metadataUri')
+            if onceux_url:
+                formats.extend(self._extract_once_formats(re.sub(
+                    r'https?://[^/]+', 'http://once.unicornmedia.com', onceux_url).replace('ads/vmap/', '')))
+
+        if not formats:
             for quality in ['sd', 'hd']:
                 # It's actually a link to a flv file
                 flv_url = streams.get('f4m_{0}'.format(quality))
@@ -71,6 +115,7 @@ class GameSpotIE(InfoExtractor):
                         'ext': 'flv',
                         'format_id': quality,
                     })
+        self._sort_formats(formats)
 
         return {
             'id': data_video['guid'],

youtube_dl/extractor/gametrailers.py

@@ -1,62 +0,0 @@
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    parse_age_limit,
-    url_basename,
-)
-
-
-class GametrailersIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.gametrailers\.com/videos/view/[^/]+/(?P<id>.+)'
-
-    _TEST = {
-        'url': 'http://www.gametrailers.com/videos/view/gametrailers-com/116437-Just-Cause-3-Review',
-        'md5': 'f28c4efa0bdfaf9b760f6507955b6a6a',
-        'info_dict': {
-            'id': '2983958',
-            'ext': 'mp4',
-            'display_id': '116437-Just-Cause-3-Review',
-            'title': 'Just Cause 3 - Review',
-            'description': 'It\'s a lot of fun to shoot at things and then watch them explode in Just Cause 3, but should there be more to the experience than that?',
-        },
-    }
-
-    def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
-        title = self._html_search_regex(
-            r'<title>(.+?)\|', webpage, 'title').strip()
-        embed_url = self._proto_relative_url(
-            self._search_regex(
-                r'src=\'(//embed.gametrailers.com/embed/[^\']+)\'', webpage,
-                'embed url'),
-            scheme='http:')
-        video_id = url_basename(embed_url)
-        embed_page = self._download_webpage(embed_url, video_id)
-        embed_vars_json = self._search_regex(
-            r'(?s)var embedVars = (\{.*?\})\s*</script>', embed_page,
-            'embed vars')
-        info = self._parse_json(embed_vars_json, video_id)
-
-        formats = []
-        for media in info['media']:
-            if media['mediaPurpose'] == 'play':
-                formats.append({
-                    'url': media['uri'],
-                    'height': media['height'],
-                    'width:': media['width'],
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'display_id': display_id,
-            'title': title,
-            'formats': formats,
-            'thumbnail': info.get('thumbUri'),
-            'description': self._og_search_description(webpage),
-            'duration': int_or_none(info.get('videoLengthInSeconds')),
-            'age_limit': parse_age_limit(info.get('audienceRating')),
-        }

youtube_dl/extractor/nbc.py

@@ -9,10 +9,6 @@ from ..utils import (
     lowercase_escape,
     smuggle_url,
     unescapeHTML,
-    update_url_query,
-    int_or_none,
-    HEADRequest,
-    parse_iso8601,
 )
 
 
@@ -192,9 +188,9 @@ class CSNNEIE(InfoExtractor):
 
 
 class NBCNewsIE(ThePlatformIE):
-    _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today)\.com/
+    _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/
         (?:video/.+?/(?P<id>\d+)|
-        ([^/]+/)*(?P<display_id>[^/?]+))
+        ([^/]+/)*(?:.*-)?(?P<mpx_id>[^/?]+))
         '''
 
     _TESTS = [
@@ -216,13 +212,16 @@ class NBCNewsIE(ThePlatformIE):
                 'ext': 'mp4',
                 'title': 'How Twitter Reacted To The Snowden Interview',
                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
+                'uploader': 'NBCU-NEWS',
+                'timestamp': 1401363060,
+                'upload_date': '20140529',
             },
         },
         {
             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
             'info_dict': {
-                'id': 'Wjf9EDR3A_60',
+                'id': '529953347624',
                 'ext': 'mp4',
                 'title': 'FULL EPISODE: Family Business',
                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
@@ -237,6 +236,9 @@ class NBCNewsIE(ThePlatformIE):
                 'ext': 'mp4',
                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
+                'timestamp': 1423104900,
+                'uploader': 'NBCU-NEWS',
+                'upload_date': '20150205',
             },
         },
         {
@@ -245,10 +247,12 @@ class NBCNewsIE(ThePlatformIE):
             'info_dict': {
                 'id': '529953347624',
                 'ext': 'mp4',
-                'title': 'Volkswagen U.S. Chief: We \'Totally Screwed Up\'',
-                'description': 'md5:d22d1281a24f22ea0880741bb4dd6301',
+                'title': 'Volkswagen U.S. Chief:\xa0 We Have Totally Screwed Up',
+                'description': 'md5:c8be487b2d80ff0594c005add88d8351',
+                'upload_date': '20150922',
+                'timestamp': 1442917800,
+                'uploader': 'NBCU-NEWS',
             },
-            'expected_warnings': ['http-6000 is not available']
         },
         {
             'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
@@ -260,6 +264,22 @@ class NBCNewsIE(ThePlatformIE):
                 'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
                 'upload_date': '20160420',
                 'timestamp': 1461152093,
+                'uploader': 'NBCU-NEWS',
+            },
+        },
+        {
+            'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
+            'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
+            'info_dict': {
+                'id': '314487875924',
+                'ext': 'mp4',
+                'title': 'The chaotic GOP immigration vote',
+                'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'timestamp': 1406937606,
+                'upload_date': '20140802',
+                'uploader': 'NBCU-NEWS',
+                'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
             },
         },
         {
@@ -290,105 +310,28 @@ class NBCNewsIE(ThePlatformIE):
             }
         else:
             # "feature" and "nightly-news" pages use theplatform.com
-            display_id = mobj.group('display_id')
-            webpage = self._download_webpage(url, display_id)
-            info = None
-            bootstrap_json = self._search_regex(
-                [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$',
-                 r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'],
-                webpage, 'bootstrap json', default=None)
-            bootstrap = self._parse_json(
-                bootstrap_json, display_id, transform_source=unescapeHTML)
-            if 'results' in bootstrap:
-                info = bootstrap['results'][0]['video']
-            elif 'video' in bootstrap:
-                info = bootstrap['video']
-            else:
-                info = bootstrap
-            video_id = info['mpxId']
-            title = info['title']
-
-            subtitles = {}
-            caption_links = info.get('captionLinks')
-            if caption_links:
-                for (sub_key, sub_ext) in (('smpte-tt', 'ttml'), ('web-vtt', 'vtt'), ('srt', 'srt')):
-                    sub_url = caption_links.get(sub_key)
-                    if sub_url:
-                        subtitles.setdefault('en', []).append({
-                            'url': sub_url,
-                            'ext': sub_ext,
-                        })
-
-            formats = []
-            for video_asset in info['videoAssets']:
-                video_url = video_asset.get('publicUrl')
-                if not video_url:
-                    continue
-                container = video_asset.get('format')
-                asset_type = video_asset.get('assetType') or ''
-                if container == 'ISM' or asset_type == 'FireTV-Once':
-                    continue
-                elif asset_type == 'OnceURL':
-                    tp_formats, tp_subtitles = self._extract_theplatform_smil(
-                        video_url, video_id)
-                    formats.extend(tp_formats)
-                    subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+            video_id = mobj.group('mpx_id')
+            if not video_id.isdigit():
+                webpage = self._download_webpage(url, video_id)
+                info = None
+                bootstrap_json = self._search_regex(
+                    [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$',
+                     r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'],
+                    webpage, 'bootstrap json', default=None)
+                bootstrap = self._parse_json(
+                    bootstrap_json, video_id, transform_source=unescapeHTML)
+                if 'results' in bootstrap:
+                    info = bootstrap['results'][0]['video']
+                elif 'video' in bootstrap:
+                    info = bootstrap['video']
                 else:
-                    tbr = int_or_none(video_asset.get('bitRate') or video_asset.get('bitrate'), 1000)
-                    format_id = 'http%s' % ('-%d' % tbr if tbr else '')
-                    video_url = update_url_query(
-                        video_url, {'format': 'redirect'})
-                    # resolve the url so that we can check availability and detect the correct extension
-                    head = self._request_webpage(
-                        HEADRequest(video_url), video_id,
-                        'Checking %s url' % format_id,
-                        '%s is not available' % format_id,
-                        fatal=False)
-                    if head:
-                        video_url = head.geturl()
-                        formats.append({
-                            'format_id': format_id,
-                            'url': video_url,
-                            'width': int_or_none(video_asset.get('width')),
-                            'height': int_or_none(video_asset.get('height')),
-                            'tbr': tbr,
-                            'container': video_asset.get('format'),
-                        })
-            self._sort_formats(formats)
+                    info = bootstrap
+                video_id = info['mpxId']
 
             return {
+                '_type': 'url_transparent',
                 'id': video_id,
-                'title': title,
-                'description': info.get('description'),
-                'thumbnail': info.get('thumbnail'),
-                'duration': int_or_none(info.get('duration')),
-                'timestamp': parse_iso8601(info.get('pubDate') or info.get('pub_date')),
-                'formats': formats,
-                'subtitles': subtitles,
+                # http://feed.theplatform.com/f/2E2eJC/nbcnews also works
+                'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byId=%s' % video_id,
+                'ie_key': 'ThePlatformFeed',
             }
-
-
-class MSNBCIE(InfoExtractor):
-    # https URLs redirect to corresponding http ones
-    _VALID_URL = r'https?://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
-    _TEST = {
-        'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
-        'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
-        'info_dict': {
-            'id': 'n_hayes_Aimm_140801_272214',
-            'ext': 'mp4',
-            'title': 'The chaotic GOP immigration vote',
-            'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'timestamp': 1406937606,
-            'upload_date': '20140802',
-            'uploader': 'NBCU-NEWS',
-            'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        embed_url = self._html_search_meta('embedURL', webpage)
-        return self.url_result(embed_url)

youtube_dl/extractor/radiojavan.py

@@ -3,7 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import(
+from ..utils import (
     unified_strdate,
     str_to_int,
 )

youtube_dl/extractor/streamcloud.py

@@ -6,7 +6,6 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
-    sanitized_Request,
     urlencode_postdata,
 )
 
@@ -45,20 +44,26 @@ class StreamcloudIE(InfoExtractor):
             (?:id="[^"]+"\s+)?
             value="([^"]*)"
             ''', orig_webpage)
-        post = urlencode_postdata(fields)
 
         self._sleep(12, video_id)
-        headers = {
-            b'Content-Type': b'application/x-www-form-urlencoded',
-        }
-        req = sanitized_Request(url, post, headers)
 
         webpage = self._download_webpage(
-            req, video_id, note='Downloading video page ...')
-        title = self._html_search_regex(
-            r'<h1[^>]*>([^<]+)<', webpage, 'title')
-        video_url = self._search_regex(
-            r'file:\s*"([^"]+)"', webpage, 'video URL')
+            url, video_id, data=urlencode_postdata(fields), headers={
+                b'Content-Type': b'application/x-www-form-urlencoded',
+            })
+
+        try:
+            title = self._html_search_regex(
+                r'<h1[^>]*>([^<]+)<', webpage, 'title')
+            video_url = self._search_regex(
+                r'file:\s*"([^"]+)"', webpage, 'video URL')
+        except ExtractorError:
+            message = self._html_search_regex(
+                r'(?s)<div[^>]+class=(["\']).*?msgboxinfo.*?\1[^>]*>(?P<message>.+?)</div>',
+                webpage, 'message', default=None, group='message')
+            if message:
+                raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+            raise
         thumbnail = self._search_regex(
             r'image:\s*"([^"]+)"', webpage, 'thumbnail URL', fatal=False)
 

youtube_dl/extractor/svt.py

@@ -6,17 +6,14 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
+    dict_get,
+    int_or_none,
+    try_get,
 )
 
 
 class SVTBaseIE(InfoExtractor):
-    def _extract_video(self, url, video_id):
-        info = self._download_json(url, video_id)
-
-        title = info['context']['title']
-        thumbnail = info['context'].get('thumbnailImage')
-
-        video_info = info['video']
+    def _extract_video(self, video_info, video_id):
         formats = []
         for vr in video_info['videoReferences']:
             player_type = vr.get('playerType')
@@ -40,27 +37,49 @@ class SVTBaseIE(InfoExtractor):
                     'format_id': player_type,
                     'url': vurl,
                 })
+        if not formats and video_info.get('rights', {}).get('geoBlockedSweden'):
+            self.raise_geo_restricted('This video is only available in Sweden')
         self._sort_formats(formats)
 
         subtitles = {}
-        subtitle_references = video_info.get('subtitleReferences')
+        subtitle_references = dict_get(video_info, ('subtitles', 'subtitleReferences'))
         if isinstance(subtitle_references, list):
             for sr in subtitle_references:
                 subtitle_url = sr.get('url')
+                subtitle_lang = sr.get('language', 'sv')
                 if subtitle_url:
-                    subtitles.setdefault('sv', []).append({'url': subtitle_url})
+                    if determine_ext(subtitle_url) == 'm3u8':
+                        # TODO(yan12125): handle WebVTT in m3u8 manifests
+                        continue
 
-        duration = video_info.get('materialLength')
-        age_limit = 18 if video_info.get('inappropriateForChildren') else 0
+                    subtitles.setdefault(subtitle_lang, []).append({'url': subtitle_url})
+
+        title = video_info.get('title')
+
+        series = video_info.get('programTitle')
+        season_number = int_or_none(video_info.get('season'))
+        episode = video_info.get('episodeTitle')
+        episode_number = int_or_none(video_info.get('episodeNumber'))
+
+        duration = int_or_none(dict_get(video_info, ('materialLength', 'contentDuration')))
+        age_limit = None
+        adult = dict_get(
+            video_info, ('inappropriateForChildren', 'blockedForChildren'),
+            skip_false_values=False)
+        if adult is not None:
+            age_limit = 18 if adult else 0
 
         return {
             'id': video_id,
             'title': title,
             'formats': formats,
             'subtitles': subtitles,
-            'thumbnail': thumbnail,
             'duration': duration,
             'age_limit': age_limit,
+            'series': series,
+            'season_number': season_number,
+            'episode': episode,
+            'episode_number': episode_number,
         }
 
 
@@ -68,11 +87,11 @@ class SVTIE(SVTBaseIE):
     _VALID_URL = r'https?://(?:www\.)?svt\.se/wd\?(?:.*?&)?widgetId=(?P<widget_id>\d+)&.*?\barticleId=(?P<id>\d+)'
     _TEST = {
         'url': 'http://www.svt.se/wd?widgetId=23991&sectionId=541&articleId=2900353&type=embed&contextSectionId=123&autostart=false',
-        'md5': '9648197555fc1b49e3dc22db4af51d46',
+        'md5': '33e9a5d8f646523ce0868ecfb0eed77d',
         'info_dict': {
             'id': '2900353',
-            'ext': 'flv',
-            'title': 'Här trycker Jagr till Giroux (under SVT-intervjun)',
+            'ext': 'mp4',
+            'title': 'Stjärnorna skojar till det - under SVT-intervjun',
             'duration': 27,
             'age_limit': 0,
         },
@@ -89,15 +108,20 @@ class SVTIE(SVTBaseIE):
         mobj = re.match(self._VALID_URL, url)
         widget_id = mobj.group('widget_id')
         article_id = mobj.group('id')
-        return self._extract_video(
+
+        info = self._download_json(
             'http://www.svt.se/wd?widgetId=%s&articleId=%s&format=json&type=embed&output=json' % (widget_id, article_id),
             article_id)
 
+        info_dict = self._extract_video(info['video'], article_id)
+        info_dict['title'] = info['context']['title']
+        return info_dict
+
 
 class SVTPlayIE(SVTBaseIE):
     IE_DESC = 'SVT Play and Öppet arkiv'
-    _VALID_URL = r'https?://(?:www\.)?(?P<host>svtplay|oppetarkiv)\.se/video/(?P<id>[0-9]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/video/(?P<id>[0-9]+)'
+    _TESTS = [{
         'url': 'http://www.svtplay.se/video/5996901/flygplan-till-haile-selassie/flygplan-till-haile-selassie-2',
         'md5': '2b6704fe4a28801e1a098bbf3c5ac611',
         'info_dict': {
@@ -113,12 +137,47 @@ class SVTPlayIE(SVTBaseIE):
                 }]
             },
         },
-    }
+    }, {
+        # geo restricted to Sweden
+        'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        host = mobj.group('host')
-        return self._extract_video(
-            'http://www.%s.se/video/%s?output=json' % (host, video_id),
-            video_id)
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        data = self._parse_json(
+            self._search_regex(
+                r'root\["__svtplay"\]\s*=\s*([^;]+);',
+                webpage, 'embedded data', default='{}'),
+            video_id, fatal=False)
+
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        if data:
+            video_info = try_get(
+                data, lambda x: x['context']['dispatcher']['stores']['VideoTitlePageStore']['data']['video'],
+                dict)
+            if video_info:
+                info_dict = self._extract_video(video_info, video_id)
+                info_dict.update({
+                    'title': data['context']['dispatcher']['stores']['MetaStore']['title'],
+                    'thumbnail': thumbnail,
+                })
+                return info_dict
+
+        video_id = self._search_regex(
+            r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)',
+            webpage, 'video id', default=None)
+
+        if video_id:
+            data = self._download_json(
+                'http://www.svt.se/videoplayer-api/video/%s' % video_id, video_id)
+            info_dict = self._extract_video(data, video_id)
+            if not info_dict.get('title'):
+                info_dict['title'] = re.sub(
+                    r'\s*\|\s*.+?$', '',
+                    info_dict.get('episode') or self._og_search_title(webpage))
+            return info_dict

youtube_dl/extractor/tf1.py

@@ -48,6 +48,6 @@ class TF1IE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         wat_id = self._html_search_regex(
-            r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8}).*?\1',
+            r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8})\1',
             webpage, 'wat id', group='id')
         return self.url_result('wat:%s' % wat_id, 'Wat')

youtube_dl/extractor/theplatform.py

@@ -277,9 +277,9 @@ class ThePlatformIE(ThePlatformBaseIE):
 
 
 class ThePlatformFeedIE(ThePlatformBaseIE):
-    _URL_TEMPLATE = '%s//feed.theplatform.com/f/%s/%s?form=json&byGuid=%s'
-    _VALID_URL = r'https?://feed\.theplatform\.com/f/(?P<provider_id>[^/]+)/(?P<feed_id>[^?/]+)\?(?:[^&]+&)*byGuid=(?P<id>[a-zA-Z0-9_]+)'
-    _TEST = {
+    _URL_TEMPLATE = '%s//feed.theplatform.com/f/%s/%s?form=json&%s'
+    _VALID_URL = r'https?://feed\.theplatform\.com/f/(?P<provider_id>[^/]+)/(?P<feed_id>[^?/]+)\?(?:[^&]+&)*(?P<filter>by(?:Gui|I)d=(?P<id>[\w-]+))'
+    _TESTS = [{
         # From http://player.theplatform.com/p/7wvmTC/MSNBCEmbeddedOffSite?guid=n_hardball_5biden_140207
         'url': 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207',
         'md5': '6e32495b5073ab414471b615c5ded394',
@@ -295,32 +295,38 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
             'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'],
             'uploader': 'NBCU-NEWS',
         },
-    }
+    }]
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-
-        video_id = mobj.group('id')
-        provider_id = mobj.group('provider_id')
-        feed_id = mobj.group('feed_id')
-
-        real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, video_id)
-        feed = self._download_json(real_url, video_id)
-        entry = feed['entries'][0]
+    def _extract_feed_info(self, provider_id, feed_id, filter_query, video_id, custom_fields=None, asset_types_query={}):
+        real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, filter_query)
+        entry = self._download_json(real_url, video_id)['entries'][0]
 
         formats = []
         subtitles = {}
         first_video_id = None
         duration = None
+        asset_types = []
         for item in entry['media$content']:
-            smil_url = item['plfile$url'] + '&mbr=true'
+            smil_url = item['plfile$url']
             cur_video_id = ThePlatformIE._match_id(smil_url)
             if first_video_id is None:
                 first_video_id = cur_video_id
                 duration = float_or_none(item.get('plfile$duration'))
-            cur_formats, cur_subtitles = self._extract_theplatform_smil(smil_url, video_id, 'Downloading SMIL data for %s' % cur_video_id)
-            formats.extend(cur_formats)
-            subtitles = self._merge_subtitles(subtitles, cur_subtitles)
+            for asset_type in item['plfile$assetTypes']:
+                if asset_type in asset_types:
+                    continue
+                asset_types.append(asset_type)
+                query = {
+                    'mbr': 'true',
+                    'formats': item['plfile$format'],
+                    'assetTypes': asset_type,
+                }
+                if asset_type in asset_types_query:
+                    query.update(asset_types_query[asset_type])
+                cur_formats, cur_subtitles = self._extract_theplatform_smil(update_url_query(
+                    smil_url, query), video_id, 'Downloading SMIL data for %s' % asset_type)
+                formats.extend(cur_formats)
+                subtitles = self._merge_subtitles(subtitles, cur_subtitles)
 
         self._sort_formats(formats)
 
@@ -344,5 +350,17 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
             'timestamp': timestamp,
             'categories': categories,
         })
+        if custom_fields:
+            ret.update(custom_fields(entry))
 
         return ret
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        provider_id = mobj.group('provider_id')
+        feed_id = mobj.group('feed_id')
+        filter_query = mobj.group('filter')
+
+        return self._extract_feed_info(provider_id, feed_id, filter_query, video_id)

youtube_dl/extractor/vimeo.py

@@ -146,7 +146,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
                             \.
                         )?
                         vimeo(?P<pro>pro)?\.com/
-                        (?!channels/[^/?#]+/?(?:$|[?#])|[^/]+/review/|(?:album|ondemand)/)
+                        (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
                         (?:.*?/)?
                         (?:
                             (?:
@@ -227,8 +227,6 @@ class VimeoIE(VimeoBaseInfoExtractor):
         {
             'url': 'http://vimeo.com/channels/keypeele/75629013',
             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
-            'note': 'Video is freely available via original URL '
-                    'and protected with password when accessed via http://vimeo.com/75629013',
             'info_dict': {
                 'id': '75629013',
                 'ext': 'mp4',
@@ -272,7 +270,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
         {
             # contains original format
             'url': 'https://vimeo.com/33951933',
-            'md5': '53c688fa95a55bf4b7293d37a89c5c53',
+            'md5': '2d9f5475e0537f013d0073e812ab89e6',
             'info_dict': {
                 'id': '33951933',
                 'ext': 'mp4',
@@ -284,6 +282,29 @@ class VimeoIE(VimeoBaseInfoExtractor):
                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
             },
         },
+        {
+            # only available via https://vimeo.com/channels/tributes/6213729 and
+            # not via https://vimeo.com/6213729
+            'url': 'https://vimeo.com/channels/tributes/6213729',
+            'info_dict': {
+                'id': '6213729',
+                'ext': 'mp4',
+                'title': 'Vimeo Tribute: The Shining',
+                'uploader': 'Casey Donahue',
+                'uploader_url': 're:https?://(?:www\.)?vimeo\.com/caseydonahue',
+                'uploader_id': 'caseydonahue',
+                'upload_date': '20090821',
+                'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'expected_warnings': ['Unable to download JSON metadata'],
+        },
+        {
+            'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
+            'only_matching': True,
+        },
         {
             'url': 'https://vimeo.com/109815029',
             'note': 'Video not completely processed, "failed" seed status',
@@ -293,6 +314,10 @@ class VimeoIE(VimeoBaseInfoExtractor):
             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
             'only_matching': True,
         },
+        {
+            'url': 'https://vimeo.com/album/2632481/video/79010983',
+            'only_matching': True,
+        },
         {
             # source file returns 403: Forbidden
             'url': 'https://vimeo.com/7809605',
@@ -369,7 +394,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
         orig_url = url
         if mobj.group('pro') or mobj.group('player'):
             url = 'https://player.vimeo.com/video/' + video_id
-        else:
+        elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
             url = 'https://vimeo.com/' + video_id
 
         # Retrieve video webpage to extract further information
@@ -630,8 +655,21 @@ class VimeoChannelIE(VimeoBaseInfoExtractor):
                 webpage = self._login_list_password(page_url, list_id, webpage)
                 yield self._extract_list_title(webpage)
 
-            for video_id in re.findall(r'id="clip_(\d+?)"', webpage):
-                yield self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo')
+            # Try extracting href first since not all videos are available via
+            # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
+            clips = re.findall(
+                r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)', webpage)
+            if clips:
+                for video_id, video_url in clips:
+                    yield self.url_result(
+                        compat_urlparse.urljoin(base_url, video_url),
+                        VimeoIE.ie_key(), video_id=video_id)
+            # More relaxed fallback
+            else:
+                for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
+                    yield self.url_result(
+                        'https://vimeo.com/%s' % video_id,
+                        VimeoIE.ie_key(), video_id=video_id)
 
             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
                 break
@@ -668,7 +706,7 @@ class VimeoUserIE(VimeoChannelIE):
 
 class VimeoAlbumIE(VimeoChannelIE):
     IE_NAME = 'vimeo:album'
-    _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)'
+    _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
     _TESTS = [{
         'url': 'https://vimeo.com/album/2632481',
@@ -688,6 +726,13 @@ class VimeoAlbumIE(VimeoChannelIE):
         'params': {
             'videopassword': 'youtube-dl',
         }
+    }, {
+        'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
+        'only_matching': True,
+    }, {
+        # TODO: respect page number
+        'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
+        'only_matching': True,
     }]
 
     def _page_url(self, base_url, pagenum):

youtube_dl/extractor/vine.py

@@ -24,6 +24,7 @@ class VineIE(InfoExtractor):
             'upload_date': '20130519',
             'uploader': 'Jack Dorsey',
             'uploader_id': '76',
+            'view_count': int,
             'like_count': int,
             'comment_count': int,
             'repost_count': int,
@@ -39,6 +40,7 @@ class VineIE(InfoExtractor):
             'upload_date': '20140815',
             'uploader': 'Mars Ruiz',
             'uploader_id': '1102363502380728320',
+            'view_count': int,
             'like_count': int,
             'comment_count': int,
             'repost_count': int,
@@ -54,6 +56,7 @@ class VineIE(InfoExtractor):
             'upload_date': '20130430',
             'uploader': 'Z3k3',
             'uploader_id': '936470460173008896',
+            'view_count': int,
             'like_count': int,
             'comment_count': int,
             'repost_count': int,
@@ -71,6 +74,7 @@ class VineIE(InfoExtractor):
             'upload_date': '20150705',
             'uploader': 'Pimry_zaa',
             'uploader_id': '1135760698325307392',
+            'view_count': int,
             'like_count': int,
             'comment_count': int,
             'repost_count': int,
@@ -109,6 +113,7 @@ class VineIE(InfoExtractor):
             'upload_date': unified_strdate(data.get('created')),
             'uploader': username,
             'uploader_id': data.get('userIdStr'),
+            'view_count': int_or_none(data.get('loops', {}).get('count')),
             'like_count': int_or_none(data.get('likes', {}).get('count')),
             'comment_count': int_or_none(data.get('comments', {}).get('count')),
             'repost_count': int_or_none(data.get('reposts', {}).get('count')),

youtube_dl/extractor/vk.py

@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 
 import re
 import json
+import sys
 
 from .common import InfoExtractor
 from ..compat import compat_str
@@ -10,7 +11,6 @@ from ..utils import (
     ExtractorError,
     int_or_none,
     orderedSet,
-    sanitized_Request,
     str_to_int,
     unescapeHTML,
     unified_strdate,
@@ -190,7 +190,7 @@ class VKIE(InfoExtractor):
         if username is None:
             return
 
-        login_page = self._download_webpage(
+        login_page, url_handle = self._download_webpage_handle(
             'https://vk.com', None, 'Downloading login page')
 
         login_form = self._hidden_inputs(login_page)
@@ -200,11 +200,26 @@ class VKIE(InfoExtractor):
             'pass': password.encode('cp1251'),
         })
 
-        request = sanitized_Request(
-            'https://login.vk.com/?act=login',
-            urlencode_postdata(login_form))
+        # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
+        # and expects the first one to be set rather than second (see
+        # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
+        # As of RFC6265 the newer one cookie should be set into cookie store
+        # what actually happens.
+        # We will workaround this VK issue by resetting the remixlhk cookie to
+        # the first one manually.
+        cookies = url_handle.headers.get('Set-Cookie')
+        if sys.version_info[0] >= 3:
+            cookies = cookies.encode('iso-8859-1')
+        cookies = cookies.decode('utf-8')
+        remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
+        if remixlhk:
+            value, domain = remixlhk.groups()
+            self._set_cookie(domain, 'remixlhk', value)
+
         login_page = self._download_webpage(
-            request, None, note='Logging in as %s' % username)
+            'https://login.vk.com/?act=login', None,
+            note='Logging in as %s' % username,
+            data=urlencode_postdata(login_form))
 
         if re.search(r'onLoginFailed', login_page):
             raise ExtractorError(

youtube_dl/extractor/xnxx.py

@@ -6,17 +6,23 @@ from ..compat import compat_urllib_parse_unquote
 
 
 class XNXXIE(InfoExtractor):
-    _VALID_URL = r'^https?://(?:video|www)\.xnxx\.com/video(?P<id>[0-9]+)/(.*)'
-    _TEST = {
-        'url': 'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_',
-        'md5': '0831677e2b4761795f68d417e0b7b445',
+    _VALID_URL = r'https?://(?:video|www)\.xnxx\.com/video-?(?P<id>[0-9a-z]+)/'
+    _TESTS = [{
+        'url': 'http://www.xnxx.com/video-55awb78/skyrim_test_video',
+        'md5': 'ef7ecee5af78f8b03dca2cf31341d3a0',
         'info_dict': {
-            'id': '1135332',
+            'id': '55awb78',
             'ext': 'flv',
-            'title': 'lida » Naked Funny Actress  (5)',
+            'title': 'Skyrim Test Video',
             'age_limit': 18,
-        }
-    }
+        },
+    }, {
+        'url': 'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.xnxx.com/video-55awb78/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/jsinterp.py

@@ -131,8 +131,9 @@ class JSInterpreter(object):
             if variable in local_vars:
                 obj = local_vars[variable]
             else:
-                obj = self._objects.setdefault(
-                    variable, self.extract_object(variable))
+                if variable not in self._objects:
+                    self._objects[variable] = self.extract_object(variable)
+                obj = self._objects[variable]
 
             if arg_str is None:
                 # Member access
@@ -203,7 +204,8 @@ class JSInterpreter(object):
             argvals = tuple([
                 int(v) if v.isdigit() else local_vars[v]
                 for v in m.group('args').split(',')])
-            self._functions.setdefault(fname, self.extract_function(fname))
+            if fname not in self._functions:
+                self._functions[fname] = self.extract_function(fname)
             return self._functions[fname](argvals)
 
         raise ExtractorError('Unsupported JS expression %r' % expr)
@@ -230,7 +232,7 @@ class JSInterpreter(object):
     def extract_function(self, funcname):
         func_m = re.search(
             r'''(?x)
-                (?:function\s+%s|[{;,]%s\s*=\s*function|var\s+%s\s*=\s*function)\s*
+                (?:function\s+%s|[{;,]\s*%s\s*=\s*function|var\s+%s\s*=\s*function)\s*
                 \((?P<args>[^)]*)\)\s*
                 \{(?P<code>[^}]+)\}''' % (
                 re.escape(funcname), re.escape(funcname), re.escape(funcname)),

youtube_dl/utils.py

@@ -2852,3 +2852,12 @@ def decode_packed_codes(code):
     return re.sub(
         r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)],
         obfucasted_code)
+
+
+def parse_m3u8_attributes(attrib):
+    info = {}
+    for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib):
+        if val.startswith('"'):
+            val = val[1:-1]
+        info[key] = val
+    return info

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.06.19.1'
+__version__ = '2016.06.23.1'