release 2016.06.26

Use the HTML5 API. Only low-resolution formats available

Related: #9839

Thanks @zhangn1985 for the overall algorithm (soimort/you-get#1224)

.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 Issues with outdated version will be rejected.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 *2016.06.26*. 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 Issues with outdated version will be rejected.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 **2016.06.26**
 
 ### 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 2016.06.26
 [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

@@ -175,3 +175,4 @@ Tomáš Čech
 Déstin Reed
 Roman Tsiupa
 Artur Krysiak
+Jakub Adam Wieczorek

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 +385,7 @@
  - **MovieFap**
  - **Moviezine**
  - **MPORA**
- - **MSNBC**
+ - **MSN**
  - **MTV**
  - **mtv.de**
  - **mtviggy.com**
@@ -503,6 +502,7 @@
  - **plus.google**: Google Plus
  - **pluzz.francetv.fr**
  - **podomatic**
+ - **PolskieRadio**
  - **PornHd**
  - **PornHub**
  - **PornHubPlaylist**
@@ -738,6 +738,7 @@
  - **vh1.com**
  - **Vice**
  - **ViceShow**
+ - **Vidbit**
  - **Viddler**
  - **video.google:search**: Google Video search
  - **video.mit.edu**

setup.py

@@ -21,25 +21,37 @@ try:
     import py2exe
 except ImportError:
     if len(sys.argv) >= 2 and sys.argv[1] == 'py2exe':
-        print("Cannot import py2exe", file=sys.stderr)
+        print('Cannot import py2exe', file=sys.stderr)
         exit(1)
 
 py2exe_options = {
-    "bundle_files": 1,
-    "compressed": 1,
-    "optimize": 2,
-    "dist_dir": '.',
-    "dll_excludes": ['w9xpopen.exe', 'crypt32.dll'],
+    'bundle_files': 1,
+    'compressed': 1,
+    'optimize': 2,
+    'dist_dir': '.',
+    'dll_excludes': ['w9xpopen.exe', 'crypt32.dll'],
 }
 
+# Get the version from youtube_dl/version.py without importing the package
+exec(compile(open('youtube_dl/version.py').read(),
+             'youtube_dl/version.py', 'exec'))
+
+DESCRIPTION = 'YouTube video downloader'
+LONG_DESCRIPTION = 'Command-line program to download videos from YouTube.com and other video sites'
+
 py2exe_console = [{
-    "script": "./youtube_dl/__main__.py",
-    "dest_base": "youtube-dl",
+    'script': './youtube_dl/__main__.py',
+    'dest_base': 'youtube-dl',
+    'version': __version__,
+    'description': DESCRIPTION,
+    'comments': LONG_DESCRIPTION,
+    'product_name': 'youtube-dl',
+    'product_version': __version__,
 }]
 
 py2exe_params = {
     'console': py2exe_console,
-    'options': {"py2exe": py2exe_options},
+    'options': {'py2exe': py2exe_options},
     'zipfile': None
 }
 
@@ -72,7 +84,7 @@ else:
         params['scripts'] = ['bin/youtube-dl']
 
 class build_lazy_extractors(Command):
-    description = "Build the extractor lazy loading module"
+    description = 'Build the extractor lazy loading module'
     user_options = []
 
     def initialize_options(self):
@@ -87,16 +99,11 @@ class build_lazy_extractors(Command):
             dry_run=self.dry_run,
         )
 
-# Get the version from youtube_dl/version.py without importing the package
-exec(compile(open('youtube_dl/version.py').read(),
-             'youtube_dl/version.py', 'exec'))
-
 setup(
     name='youtube_dl',
     version=__version__,
-    description='YouTube video downloader',
-    long_description='Small command-line program to download videos from'
-    ' YouTube.com and other video sites.',
+    description=DESCRIPTION,
+    long_description=LONG_DESCRIPTION,
     url='https://github.com/rg3/youtube-dl',
     author='Ricardo Garcia',
     author_email='ytdl@yt-dl.org',
@@ -112,17 +119,17 @@ setup(
     # test_requires = ['nosetest'],
 
     classifiers=[
-        "Topic :: Multimedia :: Video",
-        "Development Status :: 5 - Production/Stable",
-        "Environment :: Console",
-        "License :: Public Domain",
-        "Programming Language :: Python :: 2.6",
-        "Programming Language :: Python :: 2.7",
-        "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.2",
-        "Programming Language :: Python :: 3.3",
-        "Programming Language :: Python :: 3.4",
-        "Programming Language :: Python :: 3.5",
+        'Topic :: Multimedia :: Video',
+        'Development Status :: 5 - Production/Stable',
+        'Environment :: Console',
+        'License :: Public Domain',
+        'Programming Language :: Python :: 2.6',
+        'Programming Language :: Python :: 2.7',
+        'Programming Language :: Python :: 3',
+        'Programming Language :: Python :: 3.2',
+        'Programming Language :: Python :: 3.3',
+        'Programming Language :: Python :: 3.4',
+        'Programming Language :: Python :: 3.5',
     ],
 
     cmdclass={'build_lazy_extractors': build_lazy_extractors},

test/test_InfoExtractor.py

@@ -11,7 +11,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from test.helper import FakeYDL
 from youtube_dl.extractor.common import InfoExtractor
 from youtube_dl.extractor import YoutubeIE, get_info_extractor
-from youtube_dl.utils import encode_data_uri, strip_jsonp, ExtractorError
+from youtube_dl.utils import encode_data_uri, strip_jsonp, ExtractorError, RegexNotFoundError
 
 
 class TestIE(InfoExtractor):
@@ -66,6 +66,11 @@ class TestInfoExtractor(unittest.TestCase):
         self.assertEqual(ie._html_search_meta('d', html), '4')
         self.assertEqual(ie._html_search_meta('e', html), '5')
         self.assertEqual(ie._html_search_meta('f', html), '6')
+        self.assertEqual(ie._html_search_meta(('a', 'b', 'c'), html), '1')
+        self.assertEqual(ie._html_search_meta(('c', 'b', 'a'), html), '3')
+        self.assertEqual(ie._html_search_meta(('z', 'x', 'c'), html), '3')
+        self.assertRaises(RegexNotFoundError, ie._html_search_meta, 'z', html, None, fatal=True)
+        self.assertRaises(RegexNotFoundError, ie._html_search_meta, ('z', 'x'), html, None, fatal=True)
 
     def test_download_json(self):
         uri = encode_data_uri(b'{"foo": "blah"}', 'application/json')

test/test_utils.py

@@ -60,11 +60,13 @@ from youtube_dl.utils import (
     timeconvert,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     unsmuggle_url,
     uppercase_escape,
     lowercase_escape,
     url_basename,
     urlencode_postdata,
+    urshift,
     update_url_query,
     version_tuple,
     xpath_with_ns,
@@ -283,8 +285,28 @@ class TestUtil(unittest.TestCase):
             '20150202')
         self.assertEqual(unified_strdate('Feb 14th 2016 5:45PM'), '20160214')
         self.assertEqual(unified_strdate('25-09-2014'), '20140925')
+        self.assertEqual(unified_strdate('27.02.2016 17:30'), '20160227')
         self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None)
 
+    def test_unified_timestamps(self):
+        self.assertEqual(unified_timestamp('December 21, 2010'), 1292889600)
+        self.assertEqual(unified_timestamp('8/7/2009'), 1247011200)
+        self.assertEqual(unified_timestamp('Dec 14, 2012'), 1355443200)
+        self.assertEqual(unified_timestamp('2012/10/11 01:56:38 +0000'), 1349920598)
+        self.assertEqual(unified_timestamp('1968 12 10'), -33436800)
+        self.assertEqual(unified_timestamp('1968-12-10'), -33436800)
+        self.assertEqual(unified_timestamp('28/01/2014 21:00:00 +0100'), 1390939200)
+        self.assertEqual(
+            unified_timestamp('11/26/2014 11:30:00 AM PST', day_first=False),
+            1417001400)
+        self.assertEqual(
+            unified_timestamp('2/2/2015 6:47:40 PM', day_first=False),
+            1422902860)
+        self.assertEqual(unified_timestamp('Feb 14th 2016 5:45PM'), 1455471900)
+        self.assertEqual(unified_timestamp('25-09-2014'), 1411603200)
+        self.assertEqual(unified_timestamp('27.02.2016 17:30'), 1456594200)
+        self.assertEqual(unified_timestamp('UNKNOWN DATE FORMAT'), None)
+
     def test_determine_ext(self):
         self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4')
         self.assertEqual(determine_ext('http://example.com/foo/bar/?download', None), None)
@@ -959,5 +981,9 @@ The first line
         self.assertRaises(ValueError, encode_base_n, 0, 70)
         self.assertRaises(ValueError, encode_base_n, 0, 60, custom_table)
 
+    def test_urshift(self):
+        self.assertEqual(urshift(3, 1), 1)
+        self.assertEqual(urshift(-3, 1), 2147483646)
+
 if __name__ == '__main__':
     unittest.main()

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/appletrailers.py

@@ -7,6 +7,8 @@ from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
     int_or_none,
+    parse_duration,
+    unified_strdate,
 )
 
 
@@ -16,7 +18,8 @@ class AppleTrailersIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
         'info_dict': {
-            'id': 'manofsteel',
+            'id': '5111',
+            'title': 'Man of Steel',
         },
         'playlist': [
             {
@@ -70,6 +73,15 @@ class AppleTrailersIE(InfoExtractor):
             'id': 'blackthorn',
         },
         'playlist_mincount': 2,
+        'expected_warnings': ['Unable to download JSON metadata'],
+    }, {
+        # json data only available from http://trailers.apple.com/trailers/feeds/data/15881.json
+        'url': 'http://trailers.apple.com/trailers/fox/kungfupanda3/',
+        'info_dict': {
+            'id': '15881',
+            'title': 'Kung Fu Panda 3',
+        },
+        'playlist_mincount': 4,
     }, {
         'url': 'http://trailers.apple.com/ca/metropole/autrui/',
         'only_matching': True,
@@ -85,6 +97,45 @@ class AppleTrailersIE(InfoExtractor):
         movie = mobj.group('movie')
         uploader_id = mobj.group('company')
 
+        webpage = self._download_webpage(url, movie)
+        film_id = self._search_regex(r"FilmId\s*=\s*'(\d+)'", webpage, 'film id')
+        film_data = self._download_json(
+            'http://trailers.apple.com/trailers/feeds/data/%s.json' % film_id,
+            film_id, fatal=False)
+
+        if film_data:
+            entries = []
+            for clip in film_data.get('clips', []):
+                clip_title = clip['title']
+
+                formats = []
+                for version, version_data in clip.get('versions', {}).items():
+                    for size, size_data in version_data.get('sizes', {}).items():
+                        src = size_data.get('src')
+                        if not src:
+                            continue
+                        formats.append({
+                            'format_id': '%s-%s' % (version, size),
+                            'url': re.sub(r'_(\d+p.mov)', r'_h\1', src),
+                            'width': int_or_none(size_data.get('width')),
+                            'height': int_or_none(size_data.get('height')),
+                            'language': version[:2],
+                        })
+                self._sort_formats(formats)
+
+                entries.append({
+                    'id': movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', clip_title).lower(),
+                    'formats': formats,
+                    'title': clip_title,
+                    'thumbnail': clip.get('screen') or clip.get('thumb'),
+                    'duration': parse_duration(clip.get('runtime') or clip.get('faded')),
+                    'upload_date': unified_strdate(clip.get('posted')),
+                    'uploader_id': uploader_id,
+                })
+
+            page_data = film_data.get('page', {})
+            return self.playlist_result(entries, film_id, page_data.get('movie_title'))
+
         playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
 
         def fix_html(s):

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,
 )
 
 
@@ -748,10 +749,12 @@ class InfoExtractor(object):
         return self._og_search_property('url', html, **kargs)
 
     def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
+        if not isinstance(name, (list, tuple)):
+            name = [name]
         if display_name is None:
-            display_name = name
+            display_name = name[0]
         return self._html_search_regex(
-            self._meta_regex(name),
+            [self._meta_regex(n) for n in name],
             html, display_name, fatal=fatal, group='content', **kwargs)
 
     def _dc_search_uploader(self, html):
@@ -875,7 +878,11 @@ class InfoExtractor(object):
                 f['ext'] = determine_ext(f['url'])
 
             if isinstance(field_preference, (list, tuple)):
-                return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference)
+                return tuple(
+                    f.get(field)
+                    if f.get(field) is not None
+                    else ('' if field == 'format_id' else -1)
+                    for field in field_preference)
 
             preference = f.get('preference')
             if preference is None:
@@ -1150,23 +1157,11 @@ 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/dcn.py

@@ -20,7 +20,7 @@ from ..utils import (
 
 
 class DCNIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?'
 
     def _real_extract(self, url):
         show_id, video_id, season_id = re.match(self._VALID_URL, url).groups()
@@ -55,30 +55,32 @@ class DCNBaseIE(InfoExtractor):
             'is_live': is_live,
         }
 
-    def _extract_video_formats(self, webpage, video_id, entry_protocol):
+    def _extract_video_formats(self, webpage, video_id, m3u8_entry_protocol):
         formats = []
-        m3u8_url = self._html_search_regex(
-            r'file\s*:\s*"([^"]+)', webpage, 'm3u8 url', fatal=False)
-        if m3u8_url:
-            formats.extend(self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=None))
-
-        rtsp_url = self._search_regex(
-            r'<a[^>]+href="(rtsp://[^"]+)"', webpage, 'rtsp url', fatal=False)
-        if rtsp_url:
-            formats.append({
-                'url': rtsp_url,
-                'format_id': 'rtsp',
-            })
-
+        format_url_base = 'http' + self._html_search_regex(
+            [
+                r'file\s*:\s*"https?(://[^"]+)/playlist.m3u8',
+                r'<a[^>]+href="rtsp(://[^"]+)"'
+            ], webpage, 'format url')
+        # TODO: Current DASH formats are broken - $Time$ pattern in
+        # <SegmentTemplate> not implemented yet
+        # formats.extend(self._extract_mpd_formats(
+        #     format_url_base + '/manifest.mpd',
+        #     video_id, mpd_id='dash', fatal=False))
+        formats.extend(self._extract_m3u8_formats(
+            format_url_base + '/playlist.m3u8', video_id, 'mp4',
+            m3u8_entry_protocol, m3u8_id='hls', fatal=False))
+        formats.extend(self._extract_f4m_formats(
+            format_url_base + '/manifest.f4m',
+            video_id, f4m_id='hds', fatal=False))
         self._sort_formats(formats)
         return formats
 
 
 class DCNVideoIE(DCNBaseIE):
     IE_NAME = 'dcn:video'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?(?:video/[^/]+|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?(?:video(?:/[^/]+)?|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)'
+    _TESTS = [{
         'url': 'http://www.dcndigital.ae/#/video/%D8%B1%D8%AD%D9%84%D8%A9-%D8%A7%D9%84%D8%B9%D9%85%D8%B1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1/17375',
         'info_dict':
         {
@@ -94,7 +96,10 @@ class DCNVideoIE(DCNBaseIE):
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }, {
+        'url': 'http://awaan.ae/video/26723981/%D8%AF%D8%A7%D8%B1-%D8%A7%D9%84%D8%B3%D9%84%D8%A7%D9%85:-%D8%AE%D9%8A%D8%B1-%D8%AF%D9%88%D8%B1-%D8%A7%D9%84%D8%A3%D9%86%D8%B5%D8%A7%D8%B1',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -120,7 +125,7 @@ class DCNVideoIE(DCNBaseIE):
 
 class DCNLiveIE(DCNBaseIE):
     IE_NAME = 'dcn:live'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?live/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?live/(?P<id>\d+)'
 
     def _real_extract(self, url):
         channel_id = self._match_id(url)
@@ -147,7 +152,7 @@ class DCNLiveIE(DCNBaseIE):
 
 class DCNSeasonIE(InfoExtractor):
     IE_NAME = 'dcn:season'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))'
     _TEST = {
         'url': 'http://dcndigital.ae/#/program/205024/%D9%85%D8%AD%D8%A7%D8%B6%D8%B1%D8%A7%D8%AA-%D8%A7%D9%84%D8%B4%D9%8A%D8%AE-%D8%A7%D9%84%D8%B4%D8%B9%D8%B1%D8%A7%D9%88%D9%8A',
         'info_dict':

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
@@ -455,6 +454,7 @@ from .motherless import MotherlessIE
 from .motorsport import MotorsportIE
 from .movieclips import MovieClipsIE
 from .moviezine import MoviezineIE
+from .msn import MSNIE
 from .mtv import (
     MTVIE,
     MTVServicesEmbeddedIE,
@@ -481,7 +481,6 @@ from .nbc import (
     NBCNewsIE,
     NBCSportsIE,
     NBCSportsVPlayerIE,
-    MSNBCIE,
 )
 from .ndr import (
     NDRIE,
@@ -608,6 +607,7 @@ from .pluralsight import (
     PluralsightCourseIE,
 )
 from .podomatic import PodomaticIE
+from .polskieradio import PolskieRadioIE
 from .porn91 import Porn91IE
 from .pornhd import PornHdIE
 from .pornhub import (
@@ -917,6 +917,7 @@ from .vice import (
     ViceIE,
     ViceShowIE,
 )
+from .vidbit import VidbitIE
 from .viddler import ViddlerIE
 from .videodetective import VideoDetectiveIE
 from .videofyme import VideofyMeIE

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/generic.py

@@ -1091,12 +1091,17 @@ class GenericIE(InfoExtractor):
         # Dailymotion Cloud video
         {
             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
-            'md5': '49444254273501a64675a7e68c502681',
+            'md5': 'dcaf23ad0c67a256f4278bce6e0bae38',
             'info_dict': {
-                'id': '5585de919473990de4bee11b',
+                'id': 'x2uy8t3',
                 'ext': 'mp4',
-                'title': 'Le débat',
+                'title': 'Sauvons les abeilles ! - Le débat',
+                'description': 'md5:d9082128b1c5277987825d684939ca26',
                 'thumbnail': 're:^https?://.*\.jpe?g$',
+                'timestamp': 1434970506,
+                'upload_date': '20150622',
+                'uploader': 'Public Sénat',
+                'uploader_id': 'xa9gza',
             }
         },
         # OnionStudios embed

youtube_dl/extractor/iqiyi.py

@@ -1,30 +1,25 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import binascii
 import hashlib
 import itertools
 import math
-import os
-import random
 import re
 import time
-import uuid
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_parse_qs,
     compat_str,
     compat_urllib_parse_urlencode,
-    compat_urllib_parse_urlparse,
 )
 from ..utils import (
     decode_packed_codes,
     ExtractorError,
+    intlist_to_bytes,
     ohdave_rsa_encrypt,
     remove_start,
-    sanitized_Request,
-    urlencode_postdata,
-    url_basename,
+    urshift,
 )
 
 
@@ -171,70 +166,21 @@ class IqiyiIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
-        'md5': '2cb594dc2781e6c941a110d8f358118b',
+        'md5': '470a6c160618577166db1a7aac5a3606',
         'info_dict': {
             'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
+            'ext': 'mp4',
             'title': '美国德州空中惊现奇异云团 酷似UFO',
-            'ext': 'f4v',
         }
     }, {
         'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
+        'md5': 'f09f0a6a59b2da66a26bf4eda669a4cc',
         'info_dict': {
             'id': 'e3f585b550a280af23c98b6cb2be19fb',
-            'title': '名侦探柯南第752集',
-        },
-        'playlist': [{
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }],
-        'params': {
-            'skip_download': True,
+            'ext': 'mp4',
+            'title': '名侦探柯南 国语版',
         },
+        'skip': 'Geo-restricted to China',
     }, {
         'url': 'http://www.iqiyi.com/w_19rt6o8t9p.html',
         'only_matching': True,
@@ -287,13 +233,6 @@ class IqiyiIE(InfoExtractor):
         ('10', 'h1'),
     ]
 
-    AUTH_API_ERRORS = {
-        # No preview available (不允许试看鉴权失败)
-        'Q00505': 'This video requires a VIP account',
-        # End of preview time (试看结束鉴权失败)
-        'Q00506': 'Needs a VIP account for full video',
-    }
-
     def _real_initialize(self):
         self._login()
 
@@ -352,177 +291,101 @@ class IqiyiIE(InfoExtractor):
 
         return True
 
-    def _authenticate_vip_video(self, api_video_url, video_id, tvid, _uuid, do_report_warning):
-        auth_params = {
-            # version and platform hard-coded in com/qiyi/player/core/model/remote/AuthenticationRemote.as
-            'version': '2.0',
-            'platform': 'b6c13e26323c537d',
-            'aid': tvid,
+    @staticmethod
+    def _gen_sc(tvid, timestamp):
+        M = [1732584193, -271733879]
+        M.extend([~M[0], ~M[1]])
+        I_table = [7, 12, 17, 22, 5, 9, 14, 20, 4, 11, 16, 23, 6, 10, 15, 21]
+        C_base = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8388608, 432]
+
+        def L(n, t):
+            if t is None:
+                t = 0
+            return trunc(((n >> 1) + (t >> 1) << 1) + (n & 1) + (t & 1))
+
+        def trunc(n):
+            n = n % 0x100000000
+            if n > 0x7fffffff:
+                n -= 0x100000000
+            return n
+
+        def transform(string, mod):
+            num = int(string, 16)
+            return (num >> 8 * (i % 4) & 255 ^ i % mod) << ((a & 3) << 3)
+
+        C = list(C_base)
+        o = list(M)
+        k = str(timestamp - 7)
+        for i in range(13):
+            a = i
+            C[a >> 2] |= ord(k[a]) << 8 * (a % 4)
+
+        for i in range(16):
+            a = i + 13
+            start = (i >> 2) * 8
+            r = '03967743b643f66763d623d637e30733'
+            C[a >> 2] |= transform(''.join(reversed(r[start:start + 8])), 7)
+
+        for i in range(16):
+            a = i + 29
+            start = (i >> 2) * 8
+            r = '7038766939776a32776a32706b337139'
+            C[a >> 2] |= transform(r[start:start + 8], 1)
+
+        for i in range(9):
+            a = i + 45
+            if i < len(tvid):
+                C[a >> 2] |= ord(tvid[i]) << 8 * (a % 4)
+
+        for a in range(64):
+            i = a
+            I = i >> 4
+            C_index = [i, 5 * i + 1, 3 * i + 5, 7 * i][I] % 16 + urshift(a, 6)
+            m = L(L(o[0], [
+                trunc(o[1] & o[2]) | trunc(~o[1] & o[3]),
+                trunc(o[3] & o[1]) | trunc(~o[3] & o[2]),
+                o[1] ^ o[2] ^ o[3],
+                o[2] ^ trunc(o[1] | ~o[3])
+            ][I]), L(
+                trunc(int(abs(math.sin(i + 1)) * 4294967296)),
+                C[C_index] if C_index < len(C) else None))
+            I = I_table[4 * I + i % 4]
+            o = [o[3],
+                 L(o[1], trunc(trunc(m << I) | urshift(m, 32 - I))),
+                 o[1],
+                 o[2]]
+
+        new_M = [L(o[0], M[0]), L(o[1], M[1]), L(o[2], M[2]), L(o[3], M[3])]
+        s = [new_M[a >> 3] >> (1 ^ a & 7) * 4 & 15 for a in range(32)]
+        return binascii.hexlify(intlist_to_bytes(s))[1::2].decode('ascii')
+
+    def get_raw_data(self, tvid, video_id):
+        tm = int(time.time() * 1000)
+
+        sc = self._gen_sc(tvid, tm)
+        params = {
+            'platForm': 'h5',
+            'rate': 1,
             'tvid': tvid,
-            'uid': '',
-            'deviceId': _uuid,
-            'playType': 'main',  # XXX: always main?
-            'filename': os.path.splitext(url_basename(api_video_url))[0],
-        }
-
-        qd_items = compat_parse_qs(compat_urllib_parse_urlparse(api_video_url).query)
-        for key, val in qd_items.items():
-            auth_params[key] = val[0]
-
-        auth_req = sanitized_Request(
-            'http://api.vip.iqiyi.com/services/ckn.action',
-            urlencode_postdata(auth_params))
-        # iQiyi server throws HTTP 405 error without the following header
-        auth_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        auth_result = self._download_json(
-            auth_req, video_id,
-            note='Downloading video authentication JSON',
-            errnote='Unable to download video authentication JSON')
-
-        code = auth_result.get('code')
-        msg = self.AUTH_API_ERRORS.get(code) or auth_result.get('msg') or code
-        if code == 'Q00506':
-            if do_report_warning:
-                self.report_warning(msg)
-            return False
-        if 'data' not in auth_result:
-            if msg is not None:
-                raise ExtractorError('%s said: %s' % (self.IE_NAME, msg), expected=True)
-            raise ExtractorError('Unexpected error from Iqiyi auth API')
-
-        return auth_result['data']
-
-    def construct_video_urls(self, data, video_id, _uuid, tvid):
-        def do_xor(x, y):
-            a = y % 3
-            if a == 1:
-                return x ^ 121
-            if a == 2:
-                return x ^ 72
-            return x ^ 103
-
-        def get_encode_code(l):
-            a = 0
-            b = l.split('-')
-            c = len(b)
-            s = ''
-            for i in range(c - 1, -1, -1):
-                a = do_xor(int(b[c - i - 1], 16), i)
-                s += chr(a)
-            return s[::-1]
-
-        def get_path_key(x, format_id, segment_index):
-            mg = ')(*&^flash@#$%a'
-            tm = self._download_json(
-                'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
-                note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
-            )['t']
-            t = str(int(math.floor(int(tm) / (600.0))))
-            return md5_text(t + mg + x)
-
-        video_urls_dict = {}
-        need_vip_warning_report = True
-        for format_item in data['vp']['tkl'][0]['vs']:
-            if 0 < int(format_item['bid']) <= 10:
-                format_id = self.get_format(format_item['bid'])
-            else:
-                continue
-
-            video_urls = []
-
-            video_urls_info = format_item['fs']
-            if not format_item['fs'][0]['l'].startswith('/'):
-                t = get_encode_code(format_item['fs'][0]['l'])
-                if t.endswith('mp4'):
-                    video_urls_info = format_item['flvs']
-
-            for segment_index, segment in enumerate(video_urls_info):
-                vl = segment['l']
-                if not vl.startswith('/'):
-                    vl = get_encode_code(vl)
-                is_vip_video = '/vip/' in vl
-                filesize = segment['b']
-                base_url = data['vp']['du'].split('/')
-                if not is_vip_video:
-                    key = get_path_key(
-                        vl.split('/')[-1].split('.')[0], format_id, segment_index)
-                    base_url.insert(-1, key)
-                base_url = '/'.join(base_url)
-                param = {
-                    'su': _uuid,
-                    'qyid': uuid.uuid4().hex,
-                    'client': '',
-                    'z': '',
-                    'bt': '',
-                    'ct': '',
-                    'tn': str(int(time.time()))
-                }
-                api_video_url = base_url + vl
-                if is_vip_video:
-                    api_video_url = api_video_url.replace('.f4v', '.hml')
-                    auth_result = self._authenticate_vip_video(
-                        api_video_url, video_id, tvid, _uuid, need_vip_warning_report)
-                    if auth_result is False:
-                        need_vip_warning_report = False
-                        break
-                    param.update({
-                        't': auth_result['t'],
-                        # cid is hard-coded in com/qiyi/player/core/player/RuntimeData.as
-                        'cid': 'afbe8fd3d73448c9',
-                        'vid': video_id,
-                        'QY00001': auth_result['u'],
-                    })
-                api_video_url += '?' if '?' not in api_video_url else '&'
-                api_video_url += compat_urllib_parse_urlencode(param)
-                js = self._download_json(
-                    api_video_url, video_id,
-                    note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
-                video_url = js['l']
-                video_urls.append(
-                    (video_url, filesize))
-
-            video_urls_dict[format_id] = video_urls
-        return video_urls_dict
-
-    def get_format(self, bid):
-        matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
-        return matched_format_ids[0] if len(matched_format_ids) else None
-
-    def get_bid(self, format_id):
-        matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
-        return matched_bids[0] if len(matched_bids) else None
-
-    def get_raw_data(self, tvid, video_id, enc_key, _uuid):
-        tm = str(int(time.time()))
-        tail = tm + tvid
-        param = {
-            'key': 'fvip',
-            'src': md5_text('youtube-dl'),
-            'tvId': tvid,
             'vid': video_id,
-            'vinfo': 1,
-            'tm': tm,
-            'enc': md5_text(enc_key + tail),
-            'qyid': _uuid,
-            'tn': random.random(),
-            # In iQiyi's flash player, um is set to 1 if there's a logged user
-            # Some 1080P formats are only available with a logged user.
-            # Here force um=1 to trick the iQiyi server
-            'um': 1,
-            'authkey': md5_text(md5_text('') + tail),
-            'k_tag': 1,
+            'cupid': 'qc_100001_100186',
+            'type': 'mp4',
+            'nolimit': 0,
+            'agenttype': 13,
+            'src': 'd846d0c32d664d32b6b54ea48997a589',
+            'sc': sc,
+            't': tm - 7,
+            '__jsT': None,
         }
 
-        api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
-            compat_urllib_parse_urlencode(param)
-        raw_data = self._download_json(api_url, video_id)
-        return raw_data
-
-    def get_enc_key(self, video_id):
-        # TODO: automatic key extraction
-        # last update at 2016-01-22 for Zombie::bite
-        enc_key = '4a1caba4b4465345366f28da7c117d20'
-        return enc_key
+        headers = {}
+        cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
+        if cn_verification_proxy:
+            headers['Ytdl-request-proxy'] = cn_verification_proxy
+        return self._download_json(
+            'http://cache.m.iqiyi.com/jp/tmts/%s/%s/' % (tvid, video_id),
+            video_id, transform_source=lambda s: remove_start(s, 'var tvInfoJs='),
+            query=params, headers=headers)
 
     def _extract_playlist(self, webpage):
         PAGE_SIZE = 50
@@ -571,58 +434,27 @@ class IqiyiIE(InfoExtractor):
             r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
         video_id = self._search_regex(
             r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
-        _uuid = uuid.uuid4().hex
 
-        enc_key = self.get_enc_key(video_id)
+        for _ in range(5):
+            raw_data = self.get_raw_data(tvid, video_id)
 
-        raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
+            if raw_data['code'] != 'A00000':
+                if raw_data['code'] == 'A00111':
+                    self.raise_geo_restricted()
+                raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
 
-        if raw_data['code'] != 'A000000':
-            raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
+            data = raw_data['data']
 
-        data = raw_data['data']
+            # iQiYi sometimes returns Ads
+            if not isinstance(data['playInfo'], dict):
+                self._sleep(5, video_id)
+                continue
 
-        title = data['vi']['vn']
+            title = data['playInfo']['an']
+            break
 
-        # generate video_urls_dict
-        video_urls_dict = self.construct_video_urls(
-            data, video_id, _uuid, tvid)
-
-        # construct info
-        entries = []
-        for format_id in video_urls_dict:
-            video_urls = video_urls_dict[format_id]
-            for i, video_url_info in enumerate(video_urls):
-                if len(entries) < i + 1:
-                    entries.append({'formats': []})
-                entries[i]['formats'].append(
-                    {
-                        'url': video_url_info[0],
-                        'filesize': video_url_info[-1],
-                        'format_id': format_id,
-                        'preference': int(self.get_bid(format_id))
-                    }
-                )
-
-        for i in range(len(entries)):
-            self._sort_formats(entries[i]['formats'])
-            entries[i].update(
-                {
-                    'id': '%s_part%d' % (video_id, i + 1),
-                    'title': title,
-                }
-            )
-
-        if len(entries) > 1:
-            info = {
-                '_type': 'multi_video',
-                'id': video_id,
-                'title': title,
-                'entries': entries,
-            }
-        else:
-            info = entries[0]
-            info['id'] = video_id
-            info['title'] = title
-
-        return info
+        return {
+            'id': video_id,
+            'title': title,
+            'url': data['m3u'],
+        }

youtube_dl/extractor/leeco.py

@@ -23,6 +23,7 @@ from ..utils import (
     sanitized_Request,
     str_or_none,
     url_basename,
+    urshift,
 )
 
 
@@ -74,15 +75,11 @@ class LeIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    @staticmethod
-    def urshift(val, n):
-        return val >> n if val >= 0 else (val + 0x100000000) >> n
-
     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
     def ror(self, param1, param2):
         _loc3_ = 0
         while _loc3_ < param2:
-            param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
+            param1 = urshift(param1, 1) + ((param1 & 1) << 31)
             _loc3_ += 1
         return param1
 

youtube_dl/extractor/mixcloud.py

@@ -102,11 +102,11 @@ class MixcloudIE(InfoExtractor):
         description = self._og_search_description(webpage)
         like_count = parse_count(self._search_regex(
             r'\bbutton-favorite[^>]+>.*?<span[^>]+class=["\']toggle-number[^>]+>\s*([^<]+)',
-            webpage, 'like count', fatal=False))
+            webpage, 'like count', default=None))
         view_count = str_to_int(self._search_regex(
             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
              r'/listeners/?">([0-9,.]+)</a>'],
-            webpage, 'play count', fatal=False))
+            webpage, 'play count', default=None))
 
         return {
             'id': track_id,

youtube_dl/extractor/msn.py

@@ -0,0 +1,119 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    determine_ext,
+    ExtractorError,
+    int_or_none,
+    unescapeHTML,
+)
+
+
+class MSNIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?msn\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/[a-z]{2}-(?P<id>[\da-zA-Z]+)'
+    _TESTS = [{
+        'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/criminal-minds-shemar-moore-shares-a-touching-goodbye-message/vp-BBqQYNE',
+        'md5': '8442f66c116cbab1ff7098f986983458',
+        'info_dict': {
+            'id': 'BBqQYNE',
+            'display_id': 'criminal-minds-shemar-moore-shares-a-touching-goodbye-message',
+            'ext': 'mp4',
+            'title': 'Criminal Minds - Shemar Moore Shares A Touching Goodbye Message',
+            'description': 'md5:e8e89b897b222eb33a6b5067a8f1bc25',
+            'duration': 104,
+            'uploader': 'CBS Entertainment',
+            'uploader_id': 'IT0X5aoJ6bJgYerJXSDCgFmYPB1__54v',
+        },
+    }, {
+        'url': 'http://www.msn.com/en-ae/news/offbeat/meet-the-nine-year-old-self-made-millionaire/ar-BBt6ZKf',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.msn.com/en-ae/video/watch/obama-a-lot-of-people-will-be-disappointed/vi-AAhxUMH',
+        'only_matching': True,
+    }, {
+        # geo restricted
+        'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/the-first-fart-makes-you-laugh-the-last-fart-makes-you-cry/vp-AAhzIBU',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = mobj.group('id', 'display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        video = self._parse_json(
+            self._search_regex(
+                r'data-metadata\s*=\s*(["\'])(?P<data>.+?)\1',
+                webpage, 'video data', default='{}', group='data'),
+            display_id, transform_source=unescapeHTML)
+
+        if not video:
+            error = unescapeHTML(self._search_regex(
+                r'data-error=(["\'])(?P<error>.+?)\1',
+                webpage, 'error', group='error'))
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
+
+        title = video['title']
+
+        formats = []
+        for file_ in video.get('videoFiles', []):
+            format_url = file_.get('url')
+            if not format_url:
+                continue
+            ext = determine_ext(format_url)
+            # .ism is not yet supported (see
+            # https://github.com/rg3/youtube-dl/issues/8118)
+            if ext == 'ism':
+                continue
+            if 'm3u8' in format_url:
+                # m3u8_native should not be used here until
+                # https://github.com/rg3/youtube-dl/issues/9913 is fixed
+                m3u8_formats = self._extract_m3u8_formats(
+                    format_url, display_id, 'mp4',
+                    m3u8_id='hls', fatal=False)
+                # Despite metadata in m3u8 all video+audio formats are
+                # actually video-only (no audio)
+                for f in m3u8_formats:
+                    if f.get('acodec') != 'none' and f.get('vcodec') != 'none':
+                        f['acodec'] = 'none'
+                formats.extend(m3u8_formats)
+            else:
+                formats.append({
+                    'url': format_url,
+                    'ext': 'mp4',
+                    'format_id': 'http',
+                    'width': int_or_none(file_.get('width')),
+                    'height': int_or_none(file_.get('height')),
+                })
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for file_ in video.get('files', []):
+            format_url = file_.get('url')
+            format_code = file_.get('formatCode')
+            if not format_url or not format_code:
+                continue
+            if compat_str(format_code) == '3100':
+                subtitles.setdefault(file_.get('culture', 'en'), []).append({
+                    'ext': determine_ext(format_url, 'ttml'),
+                    'url': format_url,
+                })
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': video.get('description'),
+            'thumbnail': video.get('headlineImage', {}).get('url'),
+            'duration': int_or_none(video.get('durationSecs')),
+            'uploader': video.get('sourceFriendly'),
+            'uploader_id': video.get('providerId'),
+            'creator': video.get('creator'),
+            'subtitles': subtitles,
+            'formats': formats,
+        }

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/polskieradio.py

@@ -0,0 +1,95 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+    compat_urllib_parse_unquote,
+)
+from ..utils import (
+    int_or_none,
+    strip_or_none,
+    unified_timestamp,
+)
+
+
+class PolskieRadioIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?polskieradio\.pl/\d+/\d+/Artykul/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943,Prof-Andrzej-Nowak-o-historii-nie-da-sie-myslec-beznamietnie',
+        'info_dict': {
+            'id': '1587943',
+            'title': 'Prof. Andrzej Nowak: o historii nie da się myśleć beznamiętnie',
+            'description': 'md5:12f954edbf3120c5e7075e17bf9fc5c5',
+        },
+        'playlist': [{
+            'md5': '2984ee6ce9046d91fc233bc1a864a09a',
+            'info_dict': {
+                'id': '1540576',
+                'ext': 'mp3',
+                'title': 'md5:d4623290d4ac983bf924061c75c23a0d',
+                'timestamp': 1456594200,
+                'upload_date': '20160227',
+                'duration': 2364,
+            },
+        }],
+    }, {
+        'url': 'http://www.polskieradio.pl/265/5217/Artykul/1635803,Euro-2016-nie-ma-miejsca-na-blad-Polacy-graja-ze-Szwajcaria-o-cwiercfinal',
+        'info_dict': {
+            'id': '1635803',
+            'title': 'Euro 2016: nie ma miejsca na błąd. Polacy grają ze Szwajcarią o ćwierćfinał',
+            'description': 'md5:01cb7d0cad58664095d72b51a1ebada2',
+        },
+        'playlist_mincount': 12,
+    }, {
+        'url': 'http://polskieradio.pl/9/305/Artykul/1632955,Bardzo-popularne-slowo-remis',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.polskieradio.pl/7/5102/Artykul/1587943',
+        'only_matching': True,
+    }, {
+        # with mp4 video
+        'url': 'http://www.polskieradio.pl/9/299/Artykul/1634903,Brexit-Leszek-Miller-swiat-sie-nie-zawali-Europa-bedzie-trwac-dalej',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        content = self._search_regex(
+            r'(?s)<div[^>]+class="audio atarticle"[^>]*>(.+?)<script>',
+            webpage, 'content')
+
+        timestamp = unified_timestamp(self._html_search_regex(
+            r'(?s)<span[^>]+id="datetime2"[^>]*>(.+?)</span>',
+            webpage, 'timestamp', fatal=False))
+
+        entries = []
+
+        media_urls = set()
+
+        for data_media in re.findall(r'<[^>]+data-media=({[^>]+})', content):
+            media = self._parse_json(data_media, playlist_id, fatal=False)
+            if not media.get('file') or not media.get('desc'):
+                continue
+            media_url = self._proto_relative_url(media['file'], 'http:')
+            if media_url in media_urls:
+                continue
+            media_urls.add(media_url)
+            entries.append({
+                'id': compat_str(media['id']),
+                'url': media_url,
+                'title': compat_urllib_parse_unquote(media['desc']),
+                'duration': int_or_none(media.get('length')),
+                'vcodec': 'none' if media.get('provider') == 'audio' else None,
+                'timestamp': timestamp,
+            })
+
+        title = self._og_search_title(webpage).strip()
+        description = strip_or_none(self._og_search_description(webpage))
+
+        return self.playlist_result(entries, playlist_id, title, description)

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|klipp)/(?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 +137,50 @@ class SVTPlayIE(SVTBaseIE):
                 }]
             },
         },
-    }
+    }, {
+        # geo restricted to Sweden
+        'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg',
+        '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/tvp.py

@@ -4,6 +4,12 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    clean_html,
+    get_element_by_attribute,
+    ExtractorError,
+)
 
 
 class TVPIE(InfoExtractor):
@@ -21,7 +27,7 @@ class TVPIE(InfoExtractor):
         },
     }, {
         'url': 'http://www.tvp.pl/there-can-be-anything-so-i-shortened-it/17916176',
-        'md5': 'c3b15ed1af288131115ff17a17c19dda',
+        'md5': 'b0005b542e5b4de643a9690326ab1257',
         'info_dict': {
             'id': '17916176',
             'ext': 'mp4',
@@ -53,6 +59,11 @@ class TVPIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://www.tvp.pl/sess/tvplayer.php?object_id=%s' % video_id, video_id)
 
+        error_massage = get_element_by_attribute('class', 'msg error', webpage)
+        if error_massage:
+            raise ExtractorError('%s said: %s' % (
+                self.IE_NAME, clean_html(error_massage)), expected=True)
+
         title = self._search_regex(
             r'name\s*:\s*([\'"])Title\1\s*,\s*value\s*:\s*\1(?P<title>.+?)\1',
             webpage, 'title', group='title')
@@ -66,24 +77,50 @@ class TVPIE(InfoExtractor):
             r"poster\s*:\s*'([^']+)'", webpage, 'thumbnail', default=None)
 
         video_url = self._search_regex(
-            r'0:{src:([\'"])(?P<url>.*?)\1', webpage, 'formats', group='url', default=None)
-        if not video_url:
+            r'0:{src:([\'"])(?P<url>.*?)\1', webpage,
+            'formats', group='url', default=None)
+        if not video_url or 'material_niedostepny.mp4' in video_url:
             video_url = self._download_json(
                 'http://www.tvp.pl/pub/stat/videofileinfo?video_id=%s' % video_id,
                 video_id)['video_url']
 
-        ext = video_url.rsplit('.', 1)[-1]
-        if ext != 'ism/manifest':
-            if '/' in ext:
-                ext = 'mp4'
+        formats = []
+        video_url_base = self._search_regex(
+            r'(https?://.+?/video)(?:\.(?:ism|f4m|m3u8)|-\d+\.mp4)',
+            video_url, 'video base url', default=None)
+        if video_url_base:
+            # TODO: Current DASH formats are broken - $Time$ pattern in
+            # <SegmentTemplate> not implemented yet
+            # formats.extend(self._extract_mpd_formats(
+            #     video_url_base + '.ism/video.mpd',
+            #     video_id, mpd_id='dash', fatal=False))
+            formats.extend(self._extract_f4m_formats(
+                video_url_base + '.ism/video.f4m',
+                video_id, f4m_id='hds', fatal=False))
+            m3u8_formats = self._extract_m3u8_formats(
+                video_url_base + '.ism/video.m3u8', video_id,
+                'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
+            self._sort_formats(m3u8_formats)
+            m3u8_formats = list(filter(
+                lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
+                m3u8_formats))
+            formats.extend(m3u8_formats)
+            for i, m3u8_format in enumerate(m3u8_formats, 2):
+                http_url = '%s-%d.mp4' % (video_url_base, i)
+                if self._is_valid_url(http_url, video_id):
+                    f = m3u8_format.copy()
+                    f.update({
+                        'url': http_url,
+                        'format_id': f['format_id'].replace('hls', 'http'),
+                        'protocol': 'http',
+                    })
+                    formats.append(f)
+        else:
             formats = [{
                 'format_id': 'direct',
                 'url': video_url,
-                'ext': ext,
+                'ext': determine_ext(video_url, 'mp4'),
             }]
-        else:
-            m3u8_url = re.sub('([^/]*)\.ism/manifest', r'\1.ism/\1.m3u8', video_url)
-            formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
 
         self._sort_formats(formats)
 

youtube_dl/extractor/vidbit.py

@@ -0,0 +1,84 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_urlparse
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    remove_end,
+    unified_strdate,
+)
+
+
+class VidbitIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?vidbit\.co/(?:watch|embed)\?.*?\bv=(?P<id>[\da-zA-Z]+)'
+    _TESTS = [{
+        'url': 'http://www.vidbit.co/watch?v=jkL2yDOEq2',
+        'md5': '1a34b7f14defe3b8fafca9796892924d',
+        'info_dict': {
+            'id': 'jkL2yDOEq2',
+            'ext': 'mp4',
+            'title': 'Intro to VidBit',
+            'description': 'md5:5e0d6142eec00b766cbf114bfd3d16b7',
+            'thumbnail': 're:https?://.*\.jpg$',
+            'upload_date': '20160618',
+            'view_count': int,
+            'comment_count': int,
+        }
+    }, {
+        'url': 'http://www.vidbit.co/embed?v=jkL2yDOEq2&auto=0&water=0',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            compat_urlparse.urljoin(url, '/watch?v=%s' % video_id), video_id)
+
+        video_url, title = [None] * 2
+
+        config = self._parse_json(self._search_regex(
+            r'(?s)\.setup\(({.+?})\);', webpage, 'setup', default='{}'),
+            video_id, transform_source=js_to_json)
+        if config:
+            if config.get('file'):
+                video_url = compat_urlparse.urljoin(url, config['file'])
+            title = config.get('title')
+
+        if not video_url:
+            video_url = compat_urlparse.urljoin(url, self._search_regex(
+                r'file\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
+                webpage, 'video URL', group='url'))
+
+        if not title:
+            title = remove_end(
+                self._html_search_regex(
+                    (r'<h1>(.+?)</h1>', r'<title>(.+?)</title>'),
+                    webpage, 'title', default=None) or self._og_search_title(webpage),
+                ' - VidBit')
+
+        description = self._html_search_meta(
+            ('description', 'og:description', 'twitter:description'),
+            webpage, 'description')
+
+        upload_date = unified_strdate(self._html_search_meta(
+            'datePublished', webpage, 'upload date'))
+
+        view_count = int_or_none(self._search_regex(
+            r'<strong>(\d+)</strong> views',
+            webpage, 'view count', fatal=False))
+        comment_count = int_or_none(self._search_regex(
+            r'id=["\']cmt_num["\'][^>]*>\((\d+)\)',
+            webpage, 'comment count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': description,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'upload_date': upload_date,
+            'view_count': view_count,
+            'comment_count': comment_count,
+        }

youtube_dl/extractor/vimeo.py

@@ -16,6 +16,7 @@ from ..utils import (
     ExtractorError,
     InAdvancePagedList,
     int_or_none,
+    NO_DEFAULT,
     RegexNotFoundError,
     sanitized_Request,
     smuggle_url,
@@ -56,6 +57,26 @@ class VimeoBaseInfoExtractor(InfoExtractor):
         self._set_vimeo_cookie('vuid', vuid)
         self._download_webpage(login_request, None, False, 'Wrong login info')
 
+    def _verify_video_password(self, url, video_id, webpage):
+        password = self._downloader.params.get('videopassword')
+        if password is None:
+            raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
+        token, vuid = self._extract_xsrft_and_vuid(webpage)
+        data = urlencode_postdata({
+            'password': password,
+            'token': token,
+        })
+        if url.startswith('http://'):
+            # vimeo only supports https now, but the user can give an http url
+            url = url.replace('http://', 'https://')
+        password_request = sanitized_Request(url + '/password', data)
+        password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        password_request.add_header('Referer', url)
+        self._set_vimeo_cookie('vuid', vuid)
+        return self._download_webpage(
+            password_request, video_id,
+            'Verifying the password', 'Wrong password')
+
     def _extract_xsrft_and_vuid(self, webpage):
         xsrft = self._search_regex(
             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
@@ -146,7 +167,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
                             \.
                         )?
                         vimeo(?P<pro>pro)?\.com/
-                        (?!channels/[^/?#]+/?(?:$|[?#])|[^/]+/review/|(?:album|ondemand)/)
+                        (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
                         (?:.*?/)?
                         (?:
                             (?:
@@ -227,8 +248,6 @@ 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 +291,7 @@ class VimeoIE(VimeoBaseInfoExtractor):
         {
             # contains original format
             'url': 'https://vimeo.com/33951933',
-            'md5': '53c688fa95a55bf4b7293d37a89c5c53',
+            'md5': '2d9f5475e0537f013d0073e812ab89e6',
             'info_dict': {
                 'id': '33951933',
                 'ext': 'mp4',
@@ -284,6 +303,29 @@ 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 +335,10 @@ 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',
@@ -319,26 +365,6 @@ class VimeoIE(VimeoBaseInfoExtractor):
         if mobj:
             return mobj.group(1)
 
-    def _verify_video_password(self, url, video_id, webpage):
-        password = self._downloader.params.get('videopassword')
-        if password is None:
-            raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
-        token, vuid = self._extract_xsrft_and_vuid(webpage)
-        data = urlencode_postdata({
-            'password': password,
-            'token': token,
-        })
-        if url.startswith('http://'):
-            # vimeo only supports https now, but the user can give an http url
-            url = url.replace('http://', 'https://')
-        password_request = sanitized_Request(url + '/password', data)
-        password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        password_request.add_header('Referer', url)
-        self._set_vimeo_cookie('vuid', vuid)
-        return self._download_webpage(
-            password_request, video_id,
-            'Verifying the password', 'Wrong password')
-
     def _verify_player_video_password(self, url, video_id):
         password = self._downloader.params.get('videopassword')
         if password is None:
@@ -369,7 +395,7 @@ 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 +656,21 @@ 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 +707,7 @@ 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 +727,13 @@ 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):
@@ -746,12 +792,39 @@ class VimeoReviewIE(VimeoBaseInfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader_id': 'user22258446',
         }
+    }, {
+        'note': 'Password protected',
+        'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
+        'info_dict': {
+            'id': '138823582',
+            'ext': 'mp4',
+            'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
+            'uploader': 'TMB',
+            'uploader_id': 'user37284429',
+        },
+        'params': {
+            'videopassword': 'holygrail',
+        },
     }]
 
+    def _real_initialize(self):
+        self._login()
+
+    def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
+        webpage = self._download_webpage(webpage_url, video_id)
+        config_url = self._html_search_regex(
+            r'data-config-url="([^"]+)"', webpage, 'config URL',
+            default=NO_DEFAULT if video_password_verified else None)
+        if config_url is None:
+            self._verify_video_password(webpage_url, video_id, webpage)
+            config_url = self._get_config_url(
+                webpage_url, video_id, video_password_verified=True)
+        return config_url
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        config = self._download_json(
-            'https://player.vimeo.com/video/%s/config' % video_id, video_id)
+        config_url = self._get_config_url(url, video_id)
+        config = self._download_json(config_url, video_id)
         info_dict = self._parse_config(config, video_id)
         self._vimeo_sort_formats(info_dict['formats'])
         info_dict['id'] = video_id

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/extractor/youtube.py

@@ -501,6 +501,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'youtube_include_dash_manifest': True,
                 'format': '141',
             },
+            'skip': 'format 141 not served anymore',
         },
         # DASH manifest with encrypted signature
         {
@@ -517,7 +518,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             },
             'params': {
                 'youtube_include_dash_manifest': True,
-                'format': '141',
+                'format': '141/bestaudio[ext=m4a]',
             },
         },
         # JS player signature function name containing $
@@ -537,7 +538,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             },
             'params': {
                 'youtube_include_dash_manifest': True,
-                'format': '141',
+                'format': '141/bestaudio[ext=m4a]',
             },
         },
         # Controversy video
@@ -618,7 +619,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
                 'license': 'Standard YouTube License',
                 'description': 'HO09  - Women -  GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
-                'uploader': 'Olympics',
+                'uploader': 'Olympic',
                 'title': 'Hockey - Women -  GER-AUS - London 2012 Olympic Games',
             },
             'params': {
@@ -671,7 +672,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
                 'uploader': 'dorappi2000',
                 'license': 'Standard YouTube License',
-                'formats': 'mincount:33',
+                'formats': 'mincount:32',
             },
         },
         # DASH manifest with segment_list
@@ -691,7 +692,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'params': {
                 'youtube_include_dash_manifest': True,
                 'format': '135',  # bestvideo
-            }
+            },
+            'skip': 'This live event has ended.',
         },
         {
             # Multifeed videos (multiple cameras), URL is for Main Camera
@@ -762,6 +764,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
             },
             'playlist_count': 2,
+            'skip': 'Not multifeed anymore',
         },
         {
             'url': 'http://vid.plus/FlRa-iH7PGw',
@@ -814,6 +817,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'params': {
                 'skip_download': True,
             },
+            'skip': 'This video does not exist.',
         },
         {
             # Video licensed under Creative Commons
@@ -1331,7 +1335,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
                     (?:title|href)="([^"]+)"\s+
                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
-                    class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
+                    class="[^"]*"[^>]*>
                 [^<]+\.{3}\s*
                 </a>
             ''', r'\1', video_description)

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/socks.py

@@ -76,7 +76,7 @@ class Socks4Error(ProxyError):
 
     CODES = {
         91: 'request rejected or failed',
-        92: 'request rejected becasue SOCKS server cannot connect to identd on the client',
+        92: 'request rejected because SOCKS server cannot connect to identd on the client',
         93: 'request rejected because the client program and identd report different user-ids'
     }
 

youtube_dl/utils.py

@@ -110,6 +110,49 @@ ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙ
                         itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'],
                                         'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy')))
 
+DATE_FORMATS = (
+    '%d %B %Y',
+    '%d %b %Y',
+    '%B %d %Y',
+    '%b %d %Y',
+    '%b %dst %Y %I:%M',
+    '%b %dnd %Y %I:%M',
+    '%b %dth %Y %I:%M',
+    '%Y %m %d',
+    '%Y-%m-%d',
+    '%Y/%m/%d',
+    '%Y/%m/%d %H:%M:%S',
+    '%Y-%m-%d %H:%M:%S',
+    '%Y-%m-%d %H:%M:%S.%f',
+    '%d.%m.%Y %H:%M',
+    '%d.%m.%Y %H.%M',
+    '%Y-%m-%dT%H:%M:%SZ',
+    '%Y-%m-%dT%H:%M:%S.%fZ',
+    '%Y-%m-%dT%H:%M:%S.%f0Z',
+    '%Y-%m-%dT%H:%M:%S',
+    '%Y-%m-%dT%H:%M:%S.%f',
+    '%Y-%m-%dT%H:%M',
+)
+
+DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS)
+DATE_FORMATS_DAY_FIRST.extend([
+    '%d-%m-%Y',
+    '%d.%m.%Y',
+    '%d.%m.%y',
+    '%d/%m/%Y',
+    '%d/%m/%y',
+    '%d/%m/%Y %H:%M:%S',
+])
+
+DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS)
+DATE_FORMATS_MONTH_FIRST.extend([
+    '%m-%d-%Y',
+    '%m.%d.%Y',
+    '%m/%d/%Y',
+    '%m/%d/%y',
+    '%m/%d/%Y %H:%M:%S',
+])
+
 
 def preferredencoding():
     """Get preferred encoding.
@@ -975,6 +1018,24 @@ class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
     https_response = http_response
 
 
+def extract_timezone(date_str):
+    m = re.search(
+        r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
+        date_str)
+    if not m:
+        timezone = datetime.timedelta()
+    else:
+        date_str = date_str[:-len(m.group('tz'))]
+        if not m.group('sign'):
+            timezone = datetime.timedelta()
+        else:
+            sign = 1 if m.group('sign') == '+' else -1
+            timezone = datetime.timedelta(
+                hours=sign * int(m.group('hours')),
+                minutes=sign * int(m.group('minutes')))
+    return timezone, date_str
+
+
 def parse_iso8601(date_str, delimiter='T', timezone=None):
     """ Return a UNIX timestamp from the given date """
 
@@ -984,20 +1045,8 @@ def parse_iso8601(date_str, delimiter='T', timezone=None):
     date_str = re.sub(r'\.[0-9]+', '', date_str)
 
     if timezone is None:
-        m = re.search(
-            r'(?:Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)',
-            date_str)
-        if not m:
-            timezone = datetime.timedelta()
-        else:
-            date_str = date_str[:-len(m.group(0))]
-            if not m.group('sign'):
-                timezone = datetime.timedelta()
-            else:
-                sign = 1 if m.group('sign') == '+' else -1
-                timezone = datetime.timedelta(
-                    hours=sign * int(m.group('hours')),
-                    minutes=sign * int(m.group('minutes')))
+        timezone, date_str = extract_timezone(date_str)
+
     try:
         date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter)
         dt = datetime.datetime.strptime(date_str, date_format) - timezone
@@ -1006,6 +1055,10 @@ def parse_iso8601(date_str, delimiter='T', timezone=None):
         pass
 
 
+def date_formats(day_first=True):
+    return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST
+
+
 def unified_strdate(date_str, day_first=True):
     """Return a string with the date in the format YYYYMMDD"""
 
@@ -1014,53 +1067,11 @@ def unified_strdate(date_str, day_first=True):
     upload_date = None
     # Replace commas
     date_str = date_str.replace(',', ' ')
-    # %z (UTC offset) is only supported in python>=3.2
-    if not re.match(r'^[0-9]{1,2}-[0-9]{1,2}-[0-9]{4}$', date_str):
-        date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str)
     # Remove AM/PM + timezone
     date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
+    _, date_str = extract_timezone(date_str)
 
-    format_expressions = [
-        '%d %B %Y',
-        '%d %b %Y',
-        '%B %d %Y',
-        '%b %d %Y',
-        '%b %dst %Y %I:%M',
-        '%b %dnd %Y %I:%M',
-        '%b %dth %Y %I:%M',
-        '%Y %m %d',
-        '%Y-%m-%d',
-        '%Y/%m/%d',
-        '%Y/%m/%d %H:%M:%S',
-        '%Y-%m-%d %H:%M:%S',
-        '%Y-%m-%d %H:%M:%S.%f',
-        '%d.%m.%Y %H:%M',
-        '%d.%m.%Y %H.%M',
-        '%Y-%m-%dT%H:%M:%SZ',
-        '%Y-%m-%dT%H:%M:%S.%fZ',
-        '%Y-%m-%dT%H:%M:%S.%f0Z',
-        '%Y-%m-%dT%H:%M:%S',
-        '%Y-%m-%dT%H:%M:%S.%f',
-        '%Y-%m-%dT%H:%M',
-    ]
-    if day_first:
-        format_expressions.extend([
-            '%d-%m-%Y',
-            '%d.%m.%Y',
-            '%d.%m.%y',
-            '%d/%m/%Y',
-            '%d/%m/%y',
-            '%d/%m/%Y %H:%M:%S',
-        ])
-    else:
-        format_expressions.extend([
-            '%m-%d-%Y',
-            '%m.%d.%Y',
-            '%m/%d/%Y',
-            '%m/%d/%y',
-            '%m/%d/%Y %H:%M:%S',
-        ])
-    for expression in format_expressions:
+    for expression in date_formats(day_first):
         try:
             upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
         except ValueError:
@@ -1076,6 +1087,29 @@ def unified_strdate(date_str, day_first=True):
         return compat_str(upload_date)
 
 
+def unified_timestamp(date_str, day_first=True):
+    if date_str is None:
+        return None
+
+    date_str = date_str.replace(',', ' ')
+
+    pm_delta = datetime.timedelta(hours=12 if re.search(r'(?i)PM', date_str) else 0)
+    timezone, date_str = extract_timezone(date_str)
+
+    # Remove AM/PM + timezone
+    date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
+
+    for expression in date_formats(day_first):
+        try:
+            dt = datetime.datetime.strptime(date_str, expression) - timezone + pm_delta
+            return calendar.timegm(dt.timetuple())
+        except ValueError:
+            pass
+    timetuple = email.utils.parsedate_tz(date_str)
+    if timetuple:
+        return calendar.timegm(timetuple.timetuple())
+
+
 def determine_ext(url, default_ext='unknown_video'):
     if url is None:
         return default_ext
@@ -1626,6 +1660,10 @@ def float_or_none(v, scale=1, invscale=1, default=None):
         return default
 
 
+def strip_or_none(v):
+    return None if v is None else v.strip()
+
+
 def parse_duration(s):
     if not isinstance(s, compat_basestring):
         return None
@@ -2852,3 +2890,16 @@ 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
+
+
+def urshift(val, n):
+    return val >> n if val >= 0 else (val + 0x100000000) >> n

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.06.19.1'
+__version__ = '2016.06.26'