release 2016.02.05

For consistency between python 2 and 3.

README.md

@@ -455,6 +455,8 @@ The `-o` option allows users to indicate a template for the output file names. T
  - `format_id`: The sequence will be replaced by the format code specified by `--format`.
  - `duration`: The sequence will be replaced by the length of the video in seconds.
 
+Note that some of the aforementioned sequences are not guaranteed to be present since they depend on the metadata obtained by particular extractor, such sequences will be replaced with `NA`.
+
 The current default template is `%(title)s-%(id)s.%(ext)s`.
 
 In some cases, you don't want special characters such as 中, spaces, or &, such as when transferring the downloaded filename to a Windows system or the filename through an 8bit-unsafe channel. In these cases, add the `--restrict-filenames` flag to get a shorter title:

docs/supportedsites.md

@@ -91,6 +91,7 @@
  - **Canvas**
  - **CBS**
  - **CBSNews**: CBS News
+ - **CBSNewsLiveVideo**: CBS News Live Videos
  - **CBSSports**
  - **CeskaTelevize**
  - **channel9**: Channel 9
@@ -134,6 +135,8 @@
  - **DailymotionCloud**
  - **daum.net**
  - **daum.net:clip**
+ - **daum.net:playlist**
+ - **daum.net:user**
  - **DBTV**
  - **DCN**
  - **dcn:live**

test/test_YoutubeDL.py

@@ -14,6 +14,7 @@ from test.helper import FakeYDL, assertRegexpMatches
 from youtube_dl import YoutubeDL
 from youtube_dl.compat import compat_str, compat_urllib_error
 from youtube_dl.extractor import YoutubeIE
+from youtube_dl.extractor.common import InfoExtractor
 from youtube_dl.postprocessor.common import PostProcessor
 from youtube_dl.utils import ExtractorError, match_filter_func
 
@@ -247,6 +248,17 @@ class TestFormatSelection(unittest.TestCase):
 
         def format_info(f_id):
             info = YoutubeIE._formats[f_id].copy()
+
+            # XXX: In real cases InfoExtractor._parse_mpd() fills up 'acodec'
+            # and 'vcodec', while in tests such information is incomplete since
+            # commit a6c2c24479e5f4827ceb06f64d855329c0a6f593
+            # test_YoutubeDL.test_youtube_format_selection is broken without
+            # this fix
+            if 'acodec' in info and 'vcodec' not in info:
+                info['vcodec'] = 'none'
+            elif 'vcodec' in info and 'acodec' not in info:
+                info['acodec'] = 'none'
+
             info['format_id'] = f_id
             info['url'] = 'url:' + f_id
             return info
@@ -646,6 +658,42 @@ class TestYoutubeDL(unittest.TestCase):
         ydl = YDL()
         self.assertRaises(compat_urllib_error.URLError, ydl.urlopen, 'file:///etc/passwd')
 
+    def test_do_not_override_ie_key_in_url_transparent(self):
+        ydl = YDL()
+
+        class Foo1IE(InfoExtractor):
+            _VALID_URL = r'foo1:'
+
+            def _real_extract(self, url):
+                return {
+                    '_type': 'url_transparent',
+                    'url': 'foo2:',
+                    'ie_key': 'Foo2',
+                }
+
+        class Foo2IE(InfoExtractor):
+            _VALID_URL = r'foo2:'
+
+            def _real_extract(self, url):
+                return {
+                    '_type': 'url',
+                    'url': 'foo3:',
+                    'ie_key': 'Foo3',
+                }
+
+        class Foo3IE(InfoExtractor):
+            _VALID_URL = r'foo3:'
+
+            def _real_extract(self, url):
+                return _make_result([{'url': TEST_URL}])
+
+        ydl.add_info_extractor(Foo1IE(ydl))
+        ydl.add_info_extractor(Foo2IE(ydl))
+        ydl.add_info_extractor(Foo3IE(ydl))
+        ydl.extract_info('foo1:')
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['url'], TEST_URL)
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_subtitles.py

@@ -21,7 +21,7 @@ from youtube_dl.extractor import (
     NPOIE,
     ComedyCentralIE,
     NRKTVIE,
-    RaiIE,
+    RaiTVIE,
     VikiIE,
     ThePlatformIE,
     ThePlatformFeedIE,
@@ -65,16 +65,16 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(len(subtitles.keys()), 13)
-        self.assertEqual(md5(subtitles['en']), '4cd9278a35ba2305f47354ee13472260')
-        self.assertEqual(md5(subtitles['it']), '164a51f16f260476a05b50fe4c2f161d')
-        for lang in ['it', 'fr', 'de']:
+        self.assertEqual(md5(subtitles['en']), '3cb210999d3e021bd6c7f0ea751eab06')
+        self.assertEqual(md5(subtitles['it']), '6d752b98c31f1cf8d597050c7a2cb4b5')
+        for lang in ['fr', 'de']:
             self.assertTrue(subtitles.get(lang) is not None, 'Subtitles for \'%s\' not extracted' % lang)
 
-    def test_youtube_subtitles_sbv_format(self):
+    def test_youtube_subtitles_ttml_format(self):
         self.DL.params['writesubtitles'] = True
-        self.DL.params['subtitlesformat'] = 'sbv'
+        self.DL.params['subtitlesformat'] = 'ttml'
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '13aeaa0c245a8bed9a451cb643e3ad8b')
+        self.assertEqual(md5(subtitles['en']), 'e306f8c42842f723447d9f63ad65df54')
 
     def test_youtube_subtitles_vtt_format(self):
         self.DL.params['writesubtitles'] = True
@@ -260,7 +260,7 @@ class TestNRKSubtitles(BaseTestSubtitles):
 
 class TestRaiSubtitles(BaseTestSubtitles):
     url = 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html'
-    IE = RaiIE
+    IE = RaiTVIE
 
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True

test/test_youtube_lists.py

@@ -34,7 +34,7 @@ class TestYoutubeLists(unittest.TestCase):
         ie = YoutubePlaylistIE(dl)
         # TODO find a > 100 (paginating?) videos course
         result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        entries = result['entries']
+        entries = list(result['entries'])
         self.assertEqual(YoutubeIE().extract_id(entries[0]['url']), 'j9WZyLZCBzs')
         self.assertEqual(len(entries), 25)
         self.assertEqual(YoutubeIE().extract_id(entries[-1]['url']), 'rYefUsYuEp0')

youtube_dl/YoutubeDL.py

@@ -707,7 +707,6 @@ class YoutubeDL(object):
         It will also download the videos if 'download'.
         Returns the resolved ie_result.
         """
-
         result_type = ie_result.get('_type', 'video')
 
         if result_type in ('url', 'url_transparent'):
@@ -736,7 +735,7 @@ class YoutubeDL(object):
 
             force_properties = dict(
                 (k, v) for k, v in ie_result.items() if v is not None)
-            for f in ('_type', 'url'):
+            for f in ('_type', 'url', 'ie_key'):
                 if f in force_properties:
                     del force_properties[f]
             new_result = info.copy()

youtube_dl/extractor/__init__.py

@@ -90,7 +90,10 @@ from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .canvas import CanvasIE
 from .cbs import CBSIE
-from .cbsnews import CBSNewsIE
+from .cbsnews import (
+    CBSNewsIE,
+    CBSNewsLiveVideoIE,
+)
 from .cbssports import CBSSportsIE
 from .ccc import CCCIE
 from .ceskatelevize import CeskaTelevizeIE
@@ -142,6 +145,8 @@ from .dailymotion import (
 from .daum import (
     DaumIE,
     DaumClipIE,
+    DaumPlaylistIE,
+    DaumUserIE,
 )
 from .dbtv import DBTVIE
 from .dcn import (

youtube_dl/extractor/acast.py

@@ -8,11 +8,7 @@ from ..compat import compat_str
 from ..utils import int_or_none
 
 
-class ACastBaseIE(InfoExtractor):
-    _API_BASE_URL = 'https://www.acast.com/api/'
-
-
-class ACastIE(ACastBaseIE):
+class ACastIE(InfoExtractor):
     IE_NAME = 'acast'
     _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<channel>[^/]+)/(?P<id>[^/#?]+)'
     _TEST = {
@@ -23,14 +19,19 @@ class ACastIE(ACastBaseIE):
             'ext': 'mp3',
             'title': '"Where Are You?": Taipei 101, Taiwan',
             'timestamp': 1196172000000,
-            'description': 'md5:0c5d8201dfea2b93218ea986c91eee6e',
+            'description': 'md5:a0b4ef3634e63866b542e5b1199a1a0e',
             'duration': 211,
         }
     }
 
     def _real_extract(self, url):
         channel, display_id = re.match(self._VALID_URL, url).groups()
-        cast_data = self._download_json(self._API_BASE_URL + 'channels/%s/acasts/%s/playback' % (channel, display_id), display_id)
+
+        embed_page = self._download_webpage(
+            re.sub('(?:www\.)?acast\.com', 'embedcdn.acast.com', url), display_id)
+        cast_data = self._parse_json(self._search_regex(
+            r'window\[\'acast/queries\'\]\s*=\s*([^;]+);', embed_page, 'acast data'),
+            display_id)['GetAcast/%s/%s' % (channel, display_id)]
 
         return {
             'id': compat_str(cast_data['id']),
@@ -44,7 +45,7 @@ class ACastIE(ACastBaseIE):
         }
 
 
-class ACastChannelIE(ACastBaseIE):
+class ACastChannelIE(InfoExtractor):
     IE_NAME = 'acast:channel'
     _VALID_URL = r'https?://(?:www\.)?acast\.com/(?P<id>[^/#?]+)'
     _TEST = {
@@ -56,6 +57,7 @@ class ACastChannelIE(ACastBaseIE):
         },
         'playlist_mincount': 20,
     }
+    _API_BASE_URL = 'https://www.acast.com/api/'
 
     @classmethod
     def suitable(cls, url):

youtube_dl/extractor/allocine.py

@@ -8,6 +8,8 @@ from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
     qualities,
+    unescapeHTML,
+    xpath_element,
 )
 
 
@@ -31,7 +33,7 @@ class AllocineIE(InfoExtractor):
             'id': '19540403',
             'ext': 'mp4',
             'title': 'Planes 2 Bande-annonce VF',
-            'description': 'md5:eeaffe7c2d634525e21159b93acf3b1e',
+            'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway',
             'thumbnail': 're:http://.*\.jpg',
         },
     }, {
@@ -41,7 +43,7 @@ class AllocineIE(InfoExtractor):
             'id': '19544709',
             'ext': 'mp4',
             'title': 'Dragons 2 - Bande annonce finale VF',
-            'description': 'md5:71742e3a74b0d692c7fce0dd2017a4ac',
+            'description': 'md5:601d15393ac40f249648ef000720e7e3',
             'thumbnail': 're:http://.*\.jpg',
         },
     }, {
@@ -59,14 +61,18 @@ class AllocineIE(InfoExtractor):
         if typ == 'film':
             video_id = self._search_regex(r'href="/video/player_gen_cmedia=([0-9]+).+"', webpage, 'video id')
         else:
-            player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player')
-
-            player_data = json.loads(player)
-            video_id = compat_str(player_data['refMedia'])
+            player = self._search_regex(r'data-player=\'([^\']+)\'>', webpage, 'data player', default=None)
+            if player:
+                player_data = json.loads(player)
+                video_id = compat_str(player_data['refMedia'])
+            else:
+                model = self._search_regex(r'data-model="([^"]+)">', webpage, 'data model')
+                model_data = self._parse_json(unescapeHTML(model), display_id)
+                video_id = compat_str(model_data['id'])
 
         xml = self._download_xml('http://www.allocine.fr/ws/AcVisiondataV4.ashx?media=%s' % video_id, display_id)
 
-        video = xml.find('.//AcVisionVideo').attrib
+        video = xpath_element(xml, './/AcVisionVideo').attrib
         quality = qualities(['ld', 'md', 'hd'])
 
         formats = []

youtube_dl/extractor/arte.py

@@ -13,6 +13,7 @@ from ..utils import (
     unified_strdate,
     get_element_by_attribute,
     int_or_none,
+    NO_DEFAULT,
     qualities,
 )
 
@@ -93,9 +94,18 @@ class ArteTVPlus7IE(InfoExtractor):
         json_url = self._html_search_regex(
             patterns, webpage, 'json vp url', default=None)
         if not json_url:
-            iframe_url = self._html_search_regex(
-                r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
-                webpage, 'iframe url', group='url')
+            def find_iframe_url(webpage, default=NO_DEFAULT):
+                return self._html_search_regex(
+                    r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
+                    webpage, 'iframe url', group='url', default=default)
+
+            iframe_url = find_iframe_url(webpage, None)
+            if not iframe_url:
+                embed_url = self._html_search_regex(
+                    r'arte_vp_url_oembed=\'([^\']+?)\'', webpage, 'embed url')
+                player = self._download_json(
+                    embed_url, video_id, 'Downloading player page')
+                iframe_url = find_iframe_url(player['html'])
             json_url = compat_parse_qs(
                 compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]
         return self._extract_from_json_url(json_url, video_id, lang)

youtube_dl/extractor/bpb.py

@@ -1,7 +1,13 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..utils import (
+    js_to_json,
+    determine_ext,
+)
 
 
 class BpbIE(InfoExtractor):
@@ -10,7 +16,8 @@ class BpbIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.bpb.de/mediathek/297/joachim-gauck-zu-1989-und-die-erinnerung-an-die-ddr',
-        'md5': '0792086e8e2bfbac9cdf27835d5f2093',
+        # md5 fails in Python 2.6 due to buggy server response and wrong handling of urllib2
+        'md5': 'c4f84c8a8044ca9ff68bb8441d300b3f',
         'info_dict': {
             'id': '297',
             'ext': 'mp4',
@@ -25,13 +32,26 @@ class BpbIE(InfoExtractor):
 
         title = self._html_search_regex(
             r'<h2 class="white">(.*?)</h2>', webpage, 'title')
-        video_url = self._html_search_regex(
-            r'(http://film\.bpb\.de/player/dokument_[0-9]+\.mp4)',
-            webpage, 'video URL')
+        video_info_dicts = re.findall(
+            r"({\s*src:\s*'http://film\.bpb\.de/[^}]+})", webpage)
+
+        formats = []
+        for video_info in video_info_dicts:
+            video_info = self._parse_json(video_info, video_id, transform_source=js_to_json)
+            quality = video_info['quality']
+            video_url = video_info['src']
+            formats.append({
+                'url': video_url,
+                'preference': 10 if quality == 'high' else 0,
+                'format_note': quality,
+                'format_id': '%s-%s' % (quality, determine_ext(video_url)),
+            })
+
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'url': video_url,
+            'formats': formats,
             'title': title,
             'description': self._og_search_description(webpage),
         }

youtube_dl/extractor/cbsnews.py

@@ -1,15 +1,14 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-import json
-
+from .common import InfoExtractor
 from .theplatform import ThePlatformIE
+from ..utils import parse_duration
 
 
 class CBSNewsIE(ThePlatformIE):
     IE_DESC = 'CBS News'
-    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:[^/]+/)+(?P<id>[\da-z_-]+)'
+    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
 
     _TESTS = [
         {
@@ -48,14 +47,13 @@ class CBSNewsIE(ThePlatformIE):
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
-        video_info = json.loads(self._html_search_regex(
+        video_info = self._parse_json(self._html_search_regex(
             r'(?:<ul class="media-list items" id="media-related-items"><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
-            webpage, 'video JSON info'))
+            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')
@@ -88,3 +86,41 @@ class CBSNewsIE(ThePlatformIE):
             'formats': formats,
             'subtitles': subtitles,
         }
+
+
+class CBSNewsLiveVideoIE(InfoExtractor):
+    IE_DESC = 'CBS News Live Videos'
+    _VALID_URL = r'http://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
+
+    _TEST = {
+        'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
+        'info_dict': {
+            'id': 'clinton-sanders-prepare-to-face-off-in-nh',
+            'ext': 'flv',
+            'title': 'Clinton, Sanders Prepare To Face Off In NH',
+            'duration': 334,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_info = self._parse_json(self._html_search_regex(
+            r'data-story-obj=\'({.+?})\'', webpage, 'video JSON info'), video_id)['story']
+
+        hdcore_sign = 'hdcore=3.3.1'
+        f4m_formats = self._extract_f4m_formats(video_info['url'] + '&' + hdcore_sign, video_id)
+        if f4m_formats:
+            for entry in f4m_formats:
+                # URLs without the extra param induce an 404 error
+                entry.update({'extra_param_to_segment_url': hdcore_sign})
+
+        return {
+            'id': video_id,
+            'title': video_info['headline'],
+            'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
+            'duration': parse_duration(video_info.get('segmentDur')),
+            'formats': f4m_formats,
+        }

youtube_dl/extractor/common.py

@@ -1330,6 +1330,83 @@ class InfoExtractor(object):
             })
         return entries
 
+    def _download_dash_manifest(self, dash_manifest_url, video_id, fatal=True):
+        return self._download_xml(
+            dash_manifest_url, video_id,
+            note='Downloading DASH manifest',
+            errnote='Could not download DASH manifest',
+            fatal=fatal)
+
+    def _extract_dash_manifest_formats(self, dash_manifest_url, video_id, fatal=True, namespace=None, formats_dict={}):
+        dash_doc = self._download_dash_manifest(dash_manifest_url, video_id, fatal)
+        if dash_doc is False:
+            return []
+
+        return self._parse_dash_manifest(
+            dash_doc, namespace=namespace, formats_dict=formats_dict)
+
+    def _parse_dash_manifest(self, dash_doc, namespace=None, formats_dict={}):
+        def _add_ns(path):
+            return self._xpath_ns(path, namespace)
+
+        formats = []
+        for a in dash_doc.findall('.//' + _add_ns('AdaptationSet')):
+            mime_type = a.attrib.get('mimeType')
+            for r in a.findall(_add_ns('Representation')):
+                mime_type = r.attrib.get('mimeType') or mime_type
+                url_el = r.find(_add_ns('BaseURL'))
+                if mime_type == 'text/vtt':
+                    # TODO implement WebVTT downloading
+                    pass
+                elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
+                    segment_list = r.find(_add_ns('SegmentList'))
+                    format_id = r.attrib['id']
+                    video_url = url_el.text if url_el is not None else None
+                    filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
+                    f = {
+                        'format_id': format_id,
+                        'url': video_url,
+                        'width': int_or_none(r.attrib.get('width')),
+                        'height': int_or_none(r.attrib.get('height')),
+                        'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
+                        'asr': int_or_none(r.attrib.get('audioSamplingRate')),
+                        'filesize': filesize,
+                        'fps': int_or_none(r.attrib.get('frameRate')),
+                    }
+                    if segment_list is not None:
+                        initialization_url = segment_list.find(_add_ns('Initialization')).attrib['sourceURL']
+                        f.update({
+                            'initialization_url': initialization_url,
+                            'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall(_add_ns('SegmentURL'))],
+                            'protocol': 'http_dash_segments',
+                        })
+                        if not f.get('url'):
+                            f['url'] = initialization_url
+                    try:
+                        existing_format = next(
+                            fo for fo in formats
+                            if fo['format_id'] == format_id)
+                    except StopIteration:
+                        full_info = formats_dict.get(format_id, {}).copy()
+                        full_info.update(f)
+                        codecs = r.attrib.get('codecs')
+                        if codecs:
+                            if mime_type.startswith('video/'):
+                                vcodec, acodec = codecs, 'none'
+                            else:  # mime_type.startswith('audio/')
+                                vcodec, acodec = 'none', codecs
+
+                            full_info.update({
+                                'vcodec': vcodec,
+                                'acodec': acodec,
+                            })
+                        formats.append(full_info)
+                    else:
+                        existing_format.update(f)
+                else:
+                    self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
+        return formats
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()

youtube_dl/extractor/daum.py

@@ -3,16 +3,20 @@
 from __future__ import unicode_literals
 
 import re
+import itertools
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_parse_qs,
     compat_urllib_parse,
     compat_urllib_parse_unquote,
+    compat_urlparse,
 )
 from ..utils import (
     int_or_none,
     str_to_int,
     xpath_text,
+    unescapeHTML,
 )
 
 
@@ -38,7 +42,7 @@ class DaumIE(InfoExtractor):
         'info_dict': {
             'id': '65139429',
             'ext': 'mp4',
-            'title': 'md5:a100d65d09cec246d8aa9bde7de45aed',
+            'title': '1297회, \'아빠 아들로 태어나길 잘 했어\' 민수, 감동의 눈물[아빠 어디가] 20150118',
             'description': 'md5:79794514261164ff27e36a21ad229fc5',
             'upload_date': '20150604',
             'thumbnail': 're:^https?://.*\.(?:jpg|png)',
@@ -115,6 +119,7 @@ class DaumIE(InfoExtractor):
 class DaumClipIE(InfoExtractor):
     _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/(?:clip/ClipView.(?:do|tv)|mypot/View.do)\?.*?clipid=(?P<id>\d+)'
     IE_NAME = 'daum.net:clip'
+    _URL_TEMPLATE = 'http://tvpot.daum.net/clip/ClipView.do?clipid=%s'
 
     _TESTS = [{
         'url': 'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690',
@@ -133,6 +138,10 @@ class DaumClipIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    @classmethod
+    def suitable(cls, url):
+        return False if DaumPlaylistIE.suitable(url) or DaumUserIE.suitable(url) else super(DaumClipIE, cls).suitable(url)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         clip_info = self._download_json(
@@ -143,7 +152,7 @@ class DaumClipIE(InfoExtractor):
             '_type': 'url_transparent',
             'id': video_id,
             'url': 'http://tvpot.daum.net/v/%s' % clip_info['vid'],
-            'title': clip_info['title'],
+            'title': unescapeHTML(clip_info['title']),
             'thumbnail': clip_info.get('thumb_url'),
             'description': clip_info.get('contents'),
             'duration': int_or_none(clip_info.get('duration')),
@@ -151,3 +160,139 @@ class DaumClipIE(InfoExtractor):
             'view_count': int_or_none(clip_info.get('play_count')),
             'ie_key': 'Daum',
         }
+
+
+class DaumListIE(InfoExtractor):
+    def _get_entries(self, list_id, list_id_type):
+        name = None
+        entries = []
+        for pagenum in itertools.count(1):
+            list_info = self._download_json(
+                'http://tvpot.daum.net/mypot/json/GetClipInfo.do?size=48&init=true&order=date&page=%d&%s=%s' % (
+                    pagenum, list_id_type, list_id), list_id, 'Downloading list info - %s' % pagenum)
+
+            entries.extend([
+                self.url_result(
+                    'http://tvpot.daum.net/v/%s' % clip['vid'])
+                for clip in list_info['clip_list']
+            ])
+
+            if not name:
+                name = list_info.get('playlist_bean', {}).get('name') or \
+                    list_info.get('potInfo', {}).get('name')
+
+            if not list_info.get('has_more'):
+                break
+
+        return name, entries
+
+    def _check_clip(self, url, list_id):
+        query_dict = compat_parse_qs(compat_urlparse.urlparse(url).query)
+        if 'clipid' in query_dict:
+            clip_id = query_dict['clipid'][0]
+            if self._downloader.params.get('noplaylist'):
+                self.to_screen('Downloading just video %s because of --no-playlist' % clip_id)
+                return self.url_result(DaumClipIE._URL_TEMPLATE % clip_id, 'DaumClip')
+            else:
+                self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % list_id)
+
+
+class DaumPlaylistIE(DaumListIE):
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/mypot/(?:View\.do|Top\.tv)\?.*?playlistid=(?P<id>[0-9]+)'
+    IE_NAME = 'daum.net:playlist'
+    _URL_TEMPLATE = 'http://tvpot.daum.net/mypot/View.do?playlistid=%s'
+
+    _TESTS = [{
+        'note': 'Playlist url with clipid',
+        'url': 'http://tvpot.daum.net/mypot/View.do?playlistid=6213966&clipid=73806844',
+        'info_dict': {
+            'id': '6213966',
+            'title': 'Woorissica Official',
+        },
+        'playlist_mincount': 181
+    }, {
+        'note': 'Playlist url with clipid - noplaylist',
+        'url': 'http://tvpot.daum.net/mypot/View.do?playlistid=6213966&clipid=73806844',
+        'info_dict': {
+            'id': '73806844',
+            'ext': 'mp4',
+            'title': '151017 Airport',
+            'upload_date': '20160117',
+        },
+        'params': {
+            'noplaylist': True,
+            'skip_download': True,
+        }
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if DaumUserIE.suitable(url) else super(DaumPlaylistIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+
+        clip_result = self._check_clip(url, list_id)
+        if clip_result:
+            return clip_result
+
+        name, entries = self._get_entries(list_id, 'playlistid')
+
+        return self.playlist_result(entries, list_id, name)
+
+
+class DaumUserIE(DaumListIE):
+    _VALID_URL = r'https?://(?:m\.)?tvpot\.daum\.net/mypot/(?:View|Top)\.(?:do|tv)\?.*?ownerid=(?P<id>[0-9a-zA-Z]+)'
+    IE_NAME = 'daum.net:user'
+
+    _TESTS = [{
+        'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0',
+        'info_dict': {
+            'id': 'o2scDLIVbHc0',
+            'title': '마이 리틀 텔레비전',
+        },
+        'playlist_mincount': 213
+    }, {
+        'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0&clipid=73801156',
+        'info_dict': {
+            'id': '73801156',
+            'ext': 'mp4',
+            'title': '[미공개] 김구라, 오만석이 부릅니다 \'오케피\' - 마이 리틀 텔레비전 20160116',
+            'upload_date': '20160117',
+            'description': 'md5:5e91d2d6747f53575badd24bd62b9f36'
+        },
+        'params': {
+            'noplaylist': True,
+            'skip_download': True,
+        }
+    }, {
+        'note': 'Playlist url has ownerid and playlistid, playlistid takes precedence',
+        'url': 'http://tvpot.daum.net/mypot/View.do?ownerid=o2scDLIVbHc0&playlistid=6196631',
+        'info_dict': {
+            'id': '6196631',
+            'title': '마이 리틀 텔레비전 - 20160109',
+        },
+        'playlist_count': 11
+    }, {
+        'url': 'http://tvpot.daum.net/mypot/Top.do?ownerid=o2scDLIVbHc0',
+        'only_matching': True,
+    }, {
+        'url': 'http://m.tvpot.daum.net/mypot/Top.tv?ownerid=45x1okb1If50&playlistid=3569733',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+
+        clip_result = self._check_clip(url, list_id)
+        if clip_result:
+            return clip_result
+
+        query_dict = compat_parse_qs(compat_urlparse.urlparse(url).query)
+        if 'playlistid' in query_dict:
+            playlist_id = query_dict['playlistid'][0]
+            return self.url_result(DaumPlaylistIE._URL_TEMPLATE % playlist_id, 'DaumPlaylist')
+
+        name, entries = self._get_entries(list_id, 'ownerid')
+
+        return self.playlist_result(entries, list_id, name)

youtube_dl/extractor/facebook.py

@@ -6,9 +6,11 @@ import socket
 
 from .common import InfoExtractor
 from ..compat import (
+    compat_etree_fromstring,
     compat_http_client,
     compat_urllib_error,
     compat_urllib_parse_unquote,
+    compat_urllib_parse_unquote_plus,
 )
 from ..utils import (
     error_to_compat_str,
@@ -44,6 +46,9 @@ class FacebookIE(InfoExtractor):
     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
     _NETRC_MACHINE = 'facebook'
     IE_NAME = 'facebook'
+
+    _CHROME_USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36'
+
     _TESTS = [{
         'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
         'md5': '6a40d33c0eccbb1af76cf0485a052659',
@@ -65,6 +70,16 @@ class FacebookIE(InfoExtractor):
         'expected_warnings': [
             'title'
         ]
+    }, {
+        'note': 'Video with DASH manifest',
+        'url': 'https://www.facebook.com/video.php?v=957955867617029',
+        'md5': '54706e4db4f5ad58fbad82dde1f1213f',
+        'info_dict': {
+            'id': '957955867617029',
+            'ext': 'mp4',
+            'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
+            'uploader': 'Demy de Zeeuw',
+        },
     }, {
         'url': 'https://www.facebook.com/video.php?v=10204634152394104',
         'only_matching': True,
@@ -147,8 +162,9 @@ class FacebookIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
-        webpage = self._download_webpage(url, video_id)
+        req = sanitized_Request('https://www.facebook.com/video/video.php?v=%s' % video_id)
+        req.add_header('User-Agent', self._CHROME_USER_AGENT)
+        webpage = self._download_webpage(req, video_id)
 
         video_data = None
 
@@ -197,9 +213,16 @@ class FacebookIE(InfoExtractor):
                             'url': src,
                             'preference': -10 if format_id == 'progressive' else 0,
                         })
+            dash_manifest = f[0].get('dash_manifest')
+            if dash_manifest:
+                formats.extend(self._parse_dash_manifest(
+                    compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest)),
+                    namespace='urn:mpeg:dash:schema:mpd:2011'))
         if not formats:
             raise ExtractorError('Cannot find video formats')
 
+        self._sort_formats(formats)
+
         video_title = self._html_search_regex(
             r'<h2\s+[^>]*class="uiHeaderTitle"[^>]*>([^<]*)</h2>', webpage, 'title',
             default=None)

youtube_dl/extractor/gamekings.py

@@ -6,24 +6,29 @@ from ..utils import (
     xpath_text,
     xpath_with_ns,
 )
+from .youtube import YoutubeIE
 
 
 class GamekingsIE(InfoExtractor):
-    _VALID_URL = r'http://www\.gamekings\.tv/(?:videos|nieuws)/(?P<id>[^/]+)'
+    _VALID_URL = r'http://www\.gamekings\.nl/(?:videos|nieuws)/(?P<id>[^/]+)'
     _TESTS = [{
-        'url': 'http://www.gamekings.tv/videos/phoenix-wright-ace-attorney-dual-destinies-review/',
-        # MD5 is flaky, seems to change regularly
-        # 'md5': '2f32b1f7b80fdc5cb616efb4f387f8a3',
+        # YouTube embed video
+        'url': 'http://www.gamekings.nl/videos/phoenix-wright-ace-attorney-dual-destinies-review/',
+        'md5': '5208d3a17adeaef829a7861887cb9029',
         'info_dict': {
-            'id': 'phoenix-wright-ace-attorney-dual-destinies-review',
+            'id': 'HkSQKetlGOU',
             'ext': 'mp4',
-            'title': 'Phoenix Wright: Ace Attorney \u2013 Dual Destinies Review',
-            'description': 'md5:36fd701e57e8c15ac8682a2374c99731',
+            'title': 'Phoenix Wright: Ace Attorney - Dual Destinies Review',
+            'description': 'md5:db88c0e7f47e9ea50df3271b9dc72e1d',
             'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader_id': 'UCJugRGo4STYMeFr5RoOShtQ',
+            'uploader': 'Gamekings Vault',
+            'upload_date': '20151123',
         },
+        'add_ie': ['Youtube'],
     }, {
         # vimeo video
-        'url': 'http://www.gamekings.tv/videos/the-legend-of-zelda-majoras-mask/',
+        'url': 'http://www.gamekings.nl/videos/the-legend-of-zelda-majoras-mask/',
         'md5': '12bf04dfd238e70058046937657ea68d',
         'info_dict': {
             'id': 'the-legend-of-zelda-majoras-mask',
@@ -33,7 +38,7 @@ class GamekingsIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
         },
     }, {
-        'url': 'http://www.gamekings.tv/nieuws/gamekings-extra-shelly-en-david-bereiden-zich-voor-op-de-livestream/',
+        'url': 'http://www.gamekings.nl/nieuws/gamekings-extra-shelly-en-david-bereiden-zich-voor-op-de-livestream/',
         'only_matching': True,
     }]
 
@@ -43,7 +48,11 @@ class GamekingsIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         playlist_id = self._search_regex(
-            r'gogoVideo\(\s*\d+\s*,\s*"([^"]+)', webpage, 'playlist id')
+            r'gogoVideo\([^,]+,\s*"([^"]+)', webpage, 'playlist id')
+
+        # Check if a YouTube embed is used
+        if YoutubeIE.suitable(playlist_id):
+            return self.url_result(playlist_id, ie='Youtube')
 
         playlist = self._download_xml(
             'http://www.gamekings.tv/wp-content/themes/gk2010/rss_playlist.php?id=%s' % playlist_id,

youtube_dl/extractor/generic.py

@@ -1229,19 +1229,24 @@ class GenericIE(InfoExtractor):
 
         # Check for direct link to a video
         content_type = head_response.headers.get('Content-Type', '')
-        m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
+        m = re.match(r'^(?P<type>audio|video|application(?=/(?:ogg$|(?:vnd\.apple\.|x-)?mpegurl)))/(?P<format_id>.+)$', content_type)
         if m:
             upload_date = unified_strdate(
                 head_response.headers.get('Last-Modified'))
+            formats = []
+            if m.group('format_id').endswith('mpegurl'):
+                formats = self._extract_m3u8_formats(url, video_id, 'mp4')
+            else:
+                formats = [{
+                    'format_id': m.group('format_id'),
+                    'url': url,
+                    'vcodec': 'none' if m.group('type') == 'audio' else None
+                }]
             return {
                 'id': video_id,
                 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
                 'direct': True,
-                'formats': [{
-                    'format_id': m.group('format_id'),
-                    'url': url,
-                    'vcodec': 'none' if m.group('type') == 'audio' else None
-                }],
+                'formats': formats,
                 'upload_date': upload_date,
             }
 

youtube_dl/extractor/kickstarter.py

@@ -2,12 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import smuggle_url
 
 
 class KickStarterIE(InfoExtractor):
     _VALID_URL = r'https?://www\.kickstarter\.com/projects/(?P<id>[^/]*)/.*'
     _TESTS = [{
-        'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant?ref=home_location',
+        'url': 'https://www.kickstarter.com/projects/1404461844/intersection-the-story-of-josh-grant/description',
         'md5': 'c81addca81327ffa66c642b5d8b08cab',
         'info_dict': {
             'id': '1404461844',
@@ -27,7 +28,8 @@ class KickStarterIE(InfoExtractor):
             'uploader_id': 'pebble',
             'uploader': 'Pebble Technology',
             'title': 'Pebble iOS Notifications',
-        }
+        },
+        'add_ie': ['Vimeo'],
     }, {
         'url': 'https://www.kickstarter.com/projects/1420158244/power-drive-2000/widget/video.html',
         'info_dict': {
@@ -43,7 +45,7 @@ class KickStarterIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         title = self._html_search_regex(
-            r'<title>\s*(.*?)(?:\s*&mdash; Kickstarter)?\s*</title>',
+            r'<title>\s*(.*?)(?:\s*&mdash;\s*Kickstarter)?\s*</title>',
             webpage, 'title')
         video_url = self._search_regex(
             r'data-video-url="(.*?)"',
@@ -52,7 +54,7 @@ class KickStarterIE(InfoExtractor):
             return {
                 '_type': 'url_transparent',
                 'ie_key': 'Generic',
-                'url': url,
+                'url': smuggle_url(url, {'to_generic': True}),
                 'title': title,
             }
 

youtube_dl/extractor/kuwo.py

@@ -31,6 +31,10 @@ class KuwoBaseIE(InfoExtractor):
                 (file_format['ext'], file_format.get('br', ''), song_id),
                 song_id, note='Download %s url info' % file_format['format'],
             )
+
+            if song_url == 'IPDeny':
+                raise ExtractorError('This song is blocked in this region', expected=True)
+
             if song_url.startswith('http://') or song_url.startswith('https://'):
                 formats.append({
                     'url': song_url,

youtube_dl/extractor/srgssr.py

@@ -70,14 +70,11 @@ class SRGSSRIE(InfoExtractor):
                         asset_url, media_id, 'mp4', 'm3u8_native',
                         m3u8_id=format_id, fatal=False))
                 else:
-                    ext = None
-                    if protocol == 'RTMP':
-                        ext = self._search_regex(r'([a-z0-9]+):[^/]+', asset_url, 'ext')
                     formats.append({
                         'format_id': format_id,
                         'url': asset_url,
                         'preference': preference(quality),
-                        'ext': ext,
+                        'ext': 'flv' if protocol == 'RTMP' else None,
                     })
         self._sort_formats(formats)
 

youtube_dl/extractor/vidzi.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import smuggle_url
 
 
 class VidziIE(InfoExtractor):
@@ -13,6 +14,11 @@ class VidziIE(InfoExtractor):
             'id': 'cghql9yq6emu',
             'ext': 'mp4',
             'title': 'youtube-dl test video  1\\\\2\'3/4<5\\\\6ä7↭',
+            'uploader': 'vidzi.tv',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
         },
     }
 
@@ -20,19 +26,14 @@ class VidziIE(InfoExtractor):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
-        video_host = self._html_search_regex(
-            r'id=\'vplayer\'><img src="http://(.*?)/i', webpage,
-            'video host')
-        video_hash = self._html_search_regex(
-            r'\|([a-z0-9]+)\|hls\|type', webpage, 'video_hash')
-        ext = self._html_search_regex(
-            r'\|tracks\|([a-z0-9]+)\|', webpage, 'video ext')
-        video_url = 'http://' + video_host + '/' + video_hash + '/v.' + ext
         title = self._html_search_regex(
             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
 
+        # Vidzi now uses jwplayer, which can be handled by GenericIE
         return {
+            '_type': 'url_transparent',
             'id': video_id,
             'title': title,
-            'url': video_url,
+            'url': smuggle_url(url, {'to_generic': True}),
+            'ie_key': 'Generic',
         }

youtube_dl/extractor/vk.py

@@ -321,7 +321,7 @@ class VKIE(InfoExtractor):
 class VKUserVideosIE(InfoExtractor):
     IE_NAME = 'vk:uservideos'
     IE_DESC = "VK - User's Videos"
-    _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)$'
+    _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
     _TEMPLATE_URL = 'https://vk.com/videos'
     _TESTS = [{
         'url': 'http://vk.com/videos205387401',
@@ -333,6 +333,9 @@ class VKUserVideosIE(InfoExtractor):
     }, {
         'url': 'http://vk.com/videos-77521',
         'only_matching': True,
+    }, {
+        'url': 'http://vk.com/videos-97664626?section=all',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/xuite.py

@@ -34,19 +34,20 @@ class XuiteIE(InfoExtractor):
         },
     }, {
         # Video with only one format
-        'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2',
-        'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e',
+        'url': 'http://vlog.xuite.net/play/WUxxR2xCLTI1OTI1MDk5LmZsdg==',
+        'md5': '21f7b39c009b5a4615b4463df6eb7a46',
         'info_dict': {
-            'id': '3441629',
+            'id': '25925099',
             'ext': 'mp4',
-            'title': '孫燕姿 - 眼淚成詩',
+            'title': 'BigBuckBunny_320x180',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': 217.399,
-            'timestamp': 1299383640,
-            'upload_date': '20110306',
-            'uploader': 'Valen',
-            'uploader_id': '10400126',
-            'categories': ['影視娛樂'],
+            'duration': 596.458,
+            'timestamp': 1454242500,
+            'upload_date': '20160131',
+            'uploader': 'yan12125',
+            'uploader_id': '12158353',
+            'categories': ['個人短片'],
+            'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4',
         },
     }, {
         # Video with two formats

youtube_dl/extractor/youporn.py

@@ -114,15 +114,13 @@ class YouPornIE(InfoExtractor):
             formats.append(f)
         self._sort_formats(formats)
 
-        description = self._html_search_regex(
-            r'(?s)<div[^>]+class=["\']video-description["\'][^>]*>(.+?)</div>',
-            webpage, 'description', default=None)
+        description = self._og_search_description(webpage, default=None)
         thumbnail = self._search_regex(
             r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1',
             webpage, 'thumbnail', fatal=False, group='thumbnail')
 
         uploader = self._html_search_regex(
-            r'(?s)<div[^>]+class=["\']videoInfoBy["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>',
+            r'(?s)<div[^>]+class=["\']videoInfoBy(?:\s+[^"\']+)?["\'][^>]*>\s*By:\s*</div>(.+?)</(?:a|div)>',
             webpage, 'uploader', fatal=False)
         upload_date = unified_strdate(self._html_search_regex(
             r'(?s)<div[^>]+class=["\']videoInfoTime["\'][^>]*>(.+?)</div>',

youtube_dl/extractor/youtube.py

@@ -233,7 +233,7 @@ class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
 
 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
     def _process_page(self, content):
-        for playlist_id in set(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', content)):
+        for playlist_id in orderedSet(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', content)):
             yield self.url_result(
                 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
 
@@ -316,55 +316,55 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
 
         # DASH mp4 video
-        '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
-        '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
-        '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
-        '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
-        '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
-        '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},  # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
-        '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
-        '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
-        '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
-        '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
-        '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
+        '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},  # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
+        '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
+        '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
+        '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
+        '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
 
         # Dash mp4 audio
-        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
-        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
-        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
+        '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
+        '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
+        '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
 
         # Dash webm
-        '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
-        '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
-        '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
+        '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
+        '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
+        '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
+        '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
+        '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
+        '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
+        '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
+        '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
         # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
-        '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
-        '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
-        '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
-        '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
-        '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
+        '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
+        '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
+        '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
+        '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
+        '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
 
         # Dash webm audio
-        '171': {'ext': 'webm', 'acodec': 'vorbis', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
-        '172': {'ext': 'webm', 'acodec': 'vorbis', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
+        '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
+        '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
 
         # Dash webm audio with opus inside
-        '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
-        '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
-        '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
+        '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
+        '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
+        '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
 
         # RTMP (unnamed)
         '_rtmp': {'protocol': 'rtmp'},
@@ -918,7 +918,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             if lang in sub_lang_list:
                 continue
             sub_formats = []
-            for ext in ['sbv', 'vtt', 'srt']:
+            for ext in ['ttml', 'vtt']:
                 params = compat_urllib_parse.urlencode({
                     'lang': lang,
                     'v': video_id,
@@ -1035,73 +1035,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
         return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
 
-    def _parse_dash_manifest(
-            self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
-        def decrypt_sig(mobj):
-            s = mobj.group(1)
-            dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
-            return '/signature/%s' % dec_s
-        dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
-        dash_doc = self._download_xml(
-            dash_manifest_url, video_id,
-            note='Downloading DASH manifest',
-            errnote='Could not download DASH manifest',
-            fatal=fatal)
-
-        if dash_doc is False:
-            return []
-
-        formats = []
-        for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
-            mime_type = a.attrib.get('mimeType')
-            for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
-                url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
-                if url_el is None:
-                    continue
-                if mime_type == 'text/vtt':
-                    # TODO implement WebVTT downloading
-                    pass
-                elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
-                    segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
-                    format_id = r.attrib['id']
-                    video_url = url_el.text
-                    filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
-                    f = {
-                        'format_id': format_id,
-                        'url': video_url,
-                        'width': int_or_none(r.attrib.get('width')),
-                        'height': int_or_none(r.attrib.get('height')),
-                        'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
-                        'asr': int_or_none(r.attrib.get('audioSamplingRate')),
-                        'filesize': filesize,
-                        'fps': int_or_none(r.attrib.get('frameRate')),
-                    }
-                    if segment_list is not None:
-                        f.update({
-                            'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
-                            'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
-                            'protocol': 'http_dash_segments',
-                        })
-                    try:
-                        existing_format = next(
-                            fo for fo in formats
-                            if fo['format_id'] == format_id)
-                    except StopIteration:
-                        full_info = self._formats.get(format_id, {}).copy()
-                        full_info.update(f)
-                        codecs = r.attrib.get('codecs')
-                        if codecs:
-                            if full_info.get('acodec') == 'none':
-                                full_info['vcodec'] = codecs
-                            elif full_info.get('vcodec') == 'none':
-                                full_info['acodec'] = codecs
-                        formats.append(full_info)
-                    else:
-                        existing_format.update(f)
-                else:
-                    self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
-        return formats
-
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
 
@@ -1533,8 +1466,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             for dash_manifest_url in dash_mpds:
                 dash_formats = {}
                 try:
-                    for df in self._parse_dash_manifest(
-                            video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
+                    def decrypt_sig(mobj):
+                        s = mobj.group(1)
+                        dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
+                        return '/signature/%s' % dec_s
+
+                    dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
+
+                    for df in self._extract_dash_manifest_formats(
+                            dash_manifest_url, video_id, fatal=dash_mpd_fatal,
+                            namespace='urn:mpeg:DASH:schema:MPD:2011', formats_dict=self._formats):
                         # Do not overwrite DASH format found in some previous DASH manifest
                         if df['format_id'] not in dash_formats:
                             dash_formats[df['format_id']] = df

youtube_dl/postprocessor/ffmpeg.py

@@ -391,6 +391,10 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
         for (name, value) in metadata.items():
             options.extend(['-metadata', '%s=%s' % (name, value)])
 
+        # https://github.com/rg3/youtube-dl/issues/8350
+        if info['protocol'] == 'm3u8_native' or self._downloader.params.get('hls_prefer_native', False):
+            options.extend(['-bsf:a', 'aac_adtstoasc'])
+
         self._downloader.to_screen('[ffmpeg] Adding metadata to \'%s\'' % filename)
         self.run_ffmpeg(filename, temp_filename, options)
         os.remove(encodeFilename(filename))
@@ -479,6 +483,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
             self._downloader.to_screen('[ffmpeg] There aren\'t any subtitles to convert')
             return [], info
         self._downloader.to_screen('[ffmpeg] Converting subtitles')
+        sub_filenames = []
         for lang, sub in subs.items():
             ext = sub['ext']
             if ext == new_ext:
@@ -486,6 +491,8 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
                     '[ffmpeg] Subtitle file for %s is already in the requested'
                     'format' % new_ext)
                 continue
+            old_file = subtitles_filename(filename, lang, ext)
+            sub_filenames.append(old_file)
             new_file = subtitles_filename(filename, lang, new_ext)
 
             if ext == 'dfxp' or ext == 'ttml':
@@ -493,7 +500,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
                     'You have requested to convert dfxp (TTML) subtitles into another format, '
                     'which results in style information loss')
 
-                dfxp_file = subtitles_filename(filename, lang, ext)
+                dfxp_file = old_file
                 srt_file = subtitles_filename(filename, lang, 'srt')
 
                 with io.open(dfxp_file, 'rt', encoding='utf-8') as f:
@@ -511,9 +518,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
                 if new_ext == 'srt':
                     continue
 
-            self.run_ffmpeg(
-                subtitles_filename(filename, lang, ext),
-                new_file, ['-f', new_format])
+            self.run_ffmpeg(old_file, new_file, ['-f', new_format])
 
             with io.open(new_file, 'rt', encoding='utf-8') as f:
                 subs[lang] = {
@@ -521,4 +526,4 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
                     'data': f.read(),
                 }
 
-        return [], info
+        return sub_filenames, info

youtube_dl/utils.py

@@ -2017,20 +2017,27 @@ def dfxp2srt(dfxp_data):
         'ttaf1': 'http://www.w3.org/2006/10/ttaf1',
     })
 
+    class TTMLPElementParser(object):
+        out = ''
+
+        def start(self, tag, attrib):
+            if tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'):
+                self.out += '\n'
+
+        def end(self, tag):
+            pass
+
+        def data(self, data):
+            self.out += data
+
+        def close(self):
+            return self.out.strip()
+
     def parse_node(node):
-        str_or_empty = functools.partial(str_or_none, default='')
-
-        out = str_or_empty(node.text)
-
-        for child in node:
-            if child.tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'):
-                out += '\n' + str_or_empty(child.tail)
-            elif child.tag in (_x('ttml:span'), _x('ttaf1:span'), 'span'):
-                out += str_or_empty(parse_node(child))
-            else:
-                out += str_or_empty(xml.etree.ElementTree.tostring(child))
-
-        return out
+        target = TTMLPElementParser()
+        parser = xml.etree.ElementTree.XMLParser(target=target)
+        parser.feed(xml.etree.ElementTree.tostring(node))
+        return parser.close()
 
     dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8'))
     out = []

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.01.31'
+__version__ = '2016.02.05'