Merge pull request #1 from rg3/master

test

AUTHORS

@@ -129,3 +129,4 @@ Mister Hat
 Peter Ding
 jackyzy823
 George Brighton
+Remita Amine

README.md

@@ -108,7 +108,7 @@ which means you can modify it, redistribute it or use it however you like.
     --playlist-reverse               Download playlist videos in reverse order
     --xattr-set-filesize             Set file xattribute ytdl.filesize with expected filesize (experimental)
     --hls-prefer-native              Use the native HLS downloader instead of ffmpeg (experimental)
-    --external-downloader COMMAND    Use the specified external downloader. Currently supports aria2c,curl,wget
+    --external-downloader COMMAND    Use the specified external downloader. Currently supports aria2c,curl,httpie,wget
     --external-downloader-args ARGS  Give these arguments to the external downloader
 
 ## Filesystem Options:
@@ -190,7 +190,7 @@ which means you can modify it, redistribute it or use it however you like.
     --all-formats                    Download all available video formats
     --prefer-free-formats            Prefer free video formats unless a specific one is requested
     -F, --list-formats               List all available formats
-    --youtube-skip-dash-manifest     Do not download the DASH manifest on YouTube videos
+    --youtube-skip-dash-manifest     Do not download the DASH manifests and related data on YouTube videos
     --merge-output-format FORMAT     If a merge is required (e.g. bestvideo+bestaudio), output to given container format. One of mkv, mp4, ogg, webm, flv. Ignored if no
                                      merge is required
 

docs/supportedsites.md

@@ -283,6 +283,7 @@
  - **Motherless**
  - **Motorsport**: motorsport.com
  - **MovieClips**
+ - **MovieFap**
  - **Moviezine**
  - **movshare**: MovShare
  - **MPORA**
@@ -383,6 +384,7 @@
  - **Pyvideo**
  - **qqmusic**
  - **qqmusic:album**
+ - **qqmusic:playlist**
  - **qqmusic:singer**
  - **qqmusic:toplist**
  - **QuickVid**
@@ -440,6 +442,8 @@
  - **smotri:broadcast**: Smotri.com broadcasts
  - **smotri:community**: Smotri.com community videos
  - **smotri:user**: Smotri.com user videos
+ - **SnagFilms**
+ - **SnagFilmsEmbed**
  - **Snotr**
  - **Sohu**
  - **soompi**
@@ -502,6 +506,7 @@
  - **TheOnion**
  - **ThePlatform**
  - **TheSixtyOne**
+ - **ThisAmericanLife**
  - **ThisAV**
  - **THVideo**
  - **THVideoPlaylist**
@@ -542,6 +547,7 @@
  - **twitch:stream**
  - **twitch:video**
  - **twitch:vod**
+ - **TwitterCard**
  - **Ubu**
  - **udemy**
  - **udemy:course**

youtube_dl/extractor/__init__.py

@@ -323,6 +323,7 @@ from .musicvault import MusicVaultIE
 from .muzu import MuzuTVIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE
+from .myvi import MyviIE
 from .myvideo import MyVideoIE
 from .myvidster import MyVidsterIE
 from .nationalgeographic import NationalGeographicIE
@@ -432,6 +433,7 @@ from .qqmusic import (
     QQMusicSingerIE,
     QQMusicAlbumIE,
     QQMusicToplistIE,
+    QQMusicPlaylistIE,
 )
 from .quickvid import QuickVidIE
 from .r7 import R7IE
@@ -569,6 +571,7 @@ from .tf1 import TF1IE
 from .theonion import TheOnionIE
 from .theplatform import ThePlatformIE
 from .thesixtyone import TheSixtyOneIE
+from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
 from .tinypic import TinyPicIE
 from .tlc import TlcIE, TlcDeIE
@@ -732,6 +735,7 @@ from .yandexmusic import (
     YandexMusicPlaylistIE,
 )
 from .yesjapan import YesJapanIE
+from .yinyuetai import YinYueTaiIE
 from .ynet import YnetIE
 from .youjizz import YouJizzIE
 from .youku import YoukuIE

youtube_dl/extractor/clipsyndicate.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     find_xpath_attr,
@@ -10,9 +8,9 @@ from ..utils import (
 
 
 class ClipsyndicateIE(InfoExtractor):
-    _VALID_URL = r'http://www\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)'
+    _VALID_URL = r'http://(?:chic|www)\.clipsyndicate\.com/video/play(list/\d+)?/(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.clipsyndicate.com/video/play/4629301/brick_briscoe',
         'md5': '4d7d549451bad625e0ff3d7bd56d776c',
         'info_dict': {
@@ -22,11 +20,13 @@ class ClipsyndicateIE(InfoExtractor):
             'duration': 612,
             'thumbnail': 're:^https?://.+\.jpg',
         },
-    }
+    }, {
+        'url': 'http://chic.clipsyndicate.com/video/play/5844117/shark_attack',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
         js_player = self._download_webpage(
             'http://eplayer.clipsyndicate.com/embed/player.js?va_id=%s' % video_id,
             video_id, 'Downlaoding player')

youtube_dl/extractor/common.py

@@ -846,7 +846,8 @@ class InfoExtractor(object):
 
     def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
                               entry_protocol='m3u8', preference=None,
-                              m3u8_id=None, note=None, errnote=None):
+                              m3u8_id=None, note=None, errnote=None,
+                              fatal=True):
 
         formats = [{
             'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
@@ -866,7 +867,10 @@ class InfoExtractor(object):
         m3u8_doc = self._download_webpage(
             m3u8_url, video_id,
             note=note or 'Downloading m3u8 information',
-            errnote=errnote or 'Failed to download m3u8 information')
+            errnote=errnote or 'Failed to download m3u8 information',
+            fatal=fatal)
+        if m3u8_doc is False:
+            return m3u8_doc
         last_info = None
         last_media = None
         kv_rex = re.compile(

youtube_dl/extractor/crunchyroll.py

@@ -27,7 +27,7 @@ from ..aes import (
 
 
 class CrunchyrollIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
     _NETRC_MACHINE = 'crunchyroll'
     _TESTS = [{
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
@@ -45,6 +45,22 @@ class CrunchyrollIE(InfoExtractor):
             # rtmp
             'skip_download': True,
         },
+    }, {
+        'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
+        'info_dict': {
+            'id': '589804',
+            'ext': 'flv',
+            'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
+            'description': 'md5:fe2743efedb49d279552926d0bd0cd9e',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Danny Choo Network',
+            'upload_date': '20120213',
+        },
+        'params': {
+            # rtmp
+            'skip_download': True,
+        },
+
     }, {
         'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
         'only_matching': True,
@@ -251,16 +267,17 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
             stream_quality, stream_format = self._FORMAT_IDS[fmt]
             video_format = fmt + 'p'
-            streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
+            streamdata_req = compat_urllib_request.Request(
-            # urlencode doesn't work!
+                'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
-            streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality=' + stream_quality + '&media%5Fid=' + stream_id + '&video%5Fformat=' + stream_format
+                % (stream_id, stream_format, stream_quality),
+                compat_urllib_parse.urlencode({'current_page': url}).encode('utf-8'))
             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-            streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
             streamdata = self._download_xml(
                 streamdata_req, video_id,
                 note='Downloading media info for %s' % video_format)
-            video_url = streamdata.find('./host').text
+            stream_info = streamdata.find('./{default}preload/stream_info')
-            video_play_path = streamdata.find('./file').text
+            video_url = stream_info.find('./host').text
+            video_play_path = stream_info.find('./file').text
             formats.append({
                 'url': video_url,
                 'play_path': video_play_path,

youtube_dl/extractor/dailymotion.py

@@ -254,22 +254,30 @@ class DailymotionUserIE(DailymotionPlaylistIE):
 
 
 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
-    _VALID_URL = r'http://api\.dmcloud\.net/embed/[^/]+/(?P<id>[^/?]+)'
+    _VALID_URL_PREFIX = r'http://api\.dmcloud\.net/(?:player/)?embed/'
+    _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
+    _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 
-    _TEST = {
+    _TESTS = [{
         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
         # Tested at FranceTvInfo_2
         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
         'only_matching': True,
-    }
+    }, {
+        # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
+        'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
+        'only_matching': True,
+    }]
 
     @classmethod
     def _extract_dmcloud_url(self, webpage):
-        mobj = re.search(r'<iframe[^>]+src=[\'"](http://api\.dmcloud\.net/embed/[^/]+/[^\'"]+)[\'"]', webpage)
+        mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
         if mobj:
             return mobj.group(1)
 
-        mobj = re.search(r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](http://api\.dmcloud\.net/embed/[^/]+/[^\'"]+)[\'"]', webpage)
+        mobj = re.search(
+            r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
+            webpage)
         if mobj:
             return mobj.group(1)
 

youtube_dl/extractor/drtuber.py

@@ -36,25 +36,24 @@ class DrTuberIE(InfoExtractor):
             r'<source src="([^"]+)"', webpage, 'video URL')
 
         title = self._html_search_regex(
-            [r'class="hd_title" style="[^"]+">([^<]+)</h1>', r'<title>([^<]+) - \d+'],
+            [r'<p[^>]+class="title_substrate">([^<]+)</p>', r'<title>([^<]+) - \d+'],
             webpage, 'title')
 
         thumbnail = self._html_search_regex(
             r'poster="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
 
-        like_count = str_to_int(self._html_search_regex(
+        def extract_count(id_, name):
-            r'<span id="rate_likes">\s*<img[^>]+>\s*<span>([\d,\.]+)</span>',
+            return str_to_int(self._html_search_regex(
-            webpage, 'like count', fatal=False))
+                r'<span[^>]+(?:class|id)="%s"[^>]*>([\d,\.]+)</span>' % id_,
-        dislike_count = str_to_int(self._html_search_regex(
+                webpage, '%s count' % name, fatal=False))
-            r'<span id="rate_dislikes">\s*<img[^>]+>\s*<span>([\d,\.]+)</span>',
+
-            webpage, 'like count', fatal=False))
+        like_count = extract_count('rate_likes', 'like')
-        comment_count = str_to_int(self._html_search_regex(
+        dislike_count = extract_count('rate_dislikes', 'dislike')
-            r'<span class="comments_count">([\d,\.]+)</span>',
+        comment_count = extract_count('comments_count', 'comment')
-            webpage, 'comment count', fatal=False))
 
         cats_str = self._search_regex(
-            r'<span>Categories:</span><div>(.+?)</div>', webpage, 'categories', fatal=False)
+            r'<div[^>]+class="categories_list">(.+?)</div>', webpage, 'categories', fatal=False)
         categories = [] if not cats_str else re.findall(r'<a title="([^"]+)"', cats_str)
 
         return {

youtube_dl/extractor/generic.py

@@ -37,6 +37,7 @@ from .rutv import RUTVIE
 from .tvc import TVCIE
 from .sportbox import SportBoxEmbedIE
 from .smotri import SmotriIE
+from .myvi import MyviIE
 from .condenast import CondeNastIE
 from .udn import UDNEmbedIE
 from .senateisvp import SenateISVPIE
@@ -338,6 +339,17 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        # Myvi.ru embed
+        {
+            'url': 'http://www.kinomyvi.tv/news/detail/Pervij-dublirovannij-trejler--Uzhastikov-_nOw1',
+            'info_dict': {
+                'id': 'f4dafcad-ff21-423d-89b5-146cfd89fa1e',
+                'ext': 'mp4',
+                'title': 'Ужастики, русский трейлер (2015)',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'duration': 153,
+            }
+        },
         # XHamster embed
         {
             'url': 'http://www.numisc.com/forum/showthread.php?11696-FM15-which-pumiscer-was-this-%28-vid-%29-%28-alfa-as-fuck-srx-%29&s=711f5db534502e22260dec8c5e2d66d8',
@@ -669,6 +681,18 @@ class GenericIE(InfoExtractor):
                 'title': 'John Carlson Postgame 2/25/15',
             },
         },
+        # Kaltura embed (different embed code)
+        {
+            'url': 'http://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Conference-Videos/Os-Guinness-Is-It-Fools-Talk-Unbelievable-Conference-2014',
+            'info_dict': {
+                'id': '1_a52wc67y',
+                'ext': 'flv',
+                'upload_date': '20150127',
+                'uploader_id': 'PremierMedia',
+                'timestamp': int,
+                'title': 'Os Guinness // Is It Fools Talk? // Unbelievable? Conference 2014',
+            },
+        },
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
@@ -1413,6 +1437,11 @@ class GenericIE(InfoExtractor):
         if smotri_url:
             return self.url_result(smotri_url, 'Smotri')
 
+        # Look for embedded Myvi.ru player
+        myvi_url = MyviIE._extract_url(webpage)
+        if myvi_url:
+            return self.url_result(myvi_url)
+
         # Look for embeded soundcloud player
         mobj = re.search(
             r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
@@ -1492,8 +1521,8 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'Zapiks')
 
         # Look for Kaltura embeds
-        mobj = re.search(
+        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage) or
-            r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
+                re.search(r'(?s)(["\'])(?:https?:)?//cdnapisec\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?\1.*?entry_id\s*:\s*(["\'])(?P<id>[^\2]+?)\2', webpage))
         if mobj is not None:
             return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
 

youtube_dl/extractor/gfycat.py

@@ -6,12 +6,13 @@ from ..utils import (
     int_or_none,
     float_or_none,
     qualities,
+    ExtractorError,
 )
 
 
 class GfycatIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?gfycat\.com/(?P<id>[^/?#]+)'
+    _VALID_URL = r'https?://(?:www\.)?gfycat\.com/(?:ifr/)?(?P<id>[^/?#]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://gfycat.com/DeadlyDecisiveGermanpinscher',
         'info_dict': {
             'id': 'DeadlyDecisiveGermanpinscher',
@@ -27,14 +28,33 @@ class GfycatIE(InfoExtractor):
             'categories': list,
             'age_limit': 0,
         }
+    }, {
+        'url': 'http://gfycat.com/ifr/JauntyTimelyAmazontreeboa',
+        'info_dict': {
+            'id': 'JauntyTimelyAmazontreeboa',
+            'ext': 'mp4',
+            'title': 'JauntyTimelyAmazontreeboa',
+            'timestamp': 1411720126,
+            'upload_date': '20140926',
+            'uploader': 'anonymous',
+            'duration': 3.52,
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'categories': list,
+            'age_limit': 0,
         }
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         gfy = self._download_json(
             'http://gfycat.com/cajax/get/%s' % video_id,
-            video_id, 'Downloading video info')['gfyItem']
+            video_id, 'Downloading video info')
+        if 'error' in gfy:
+            raise ExtractorError('Gfycat said: ' + gfy['error'], expected=True)
+        gfy = gfy['gfyItem']
 
         title = gfy.get('title') or gfy['gfyName']
         description = gfy.get('description')

youtube_dl/extractor/hentaistigma.py

@@ -1,7 +1,5 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 
@@ -19,20 +17,19 @@ class HentaiStigmaIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
 
         webpage = self._download_webpage(url, video_id)
 
         title = self._html_search_regex(
-            r'<h2 class="posttitle"><a[^>]*>([^<]+)</a>',
+            r'<h2[^>]+class="posttitle"[^>]*><a[^>]*>([^<]+)</a>',
             webpage, 'title')
         wrap_url = self._html_search_regex(
-            r'<iframe src="([^"]+mp4)"', webpage, 'wrapper url')
+            r'<iframe[^>]+src="([^"]+mp4)"', webpage, 'wrapper url')
         wrap_webpage = self._download_webpage(wrap_url, video_id)
 
         video_url = self._html_search_regex(
-            r'clip:\s*{\s*url: "([^"]*)"', wrap_webpage, 'video url')
+            r'file\s*:\s*"([^"]+)"', wrap_webpage, 'video url')
 
         return {
             'id': video_id,

youtube_dl/extractor/howcast.py

@@ -1,8 +1,7 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import parse_iso8601
 
 
 class HowcastIE(InfoExtractor):
@@ -13,29 +12,31 @@ class HowcastIE(InfoExtractor):
         'info_dict': {
             'id': '390161',
             'ext': 'mp4',
-            'description': 'The square knot, also known as the reef knot, is one of the oldest, most basic knots to tie, and can be used in many different ways. Here\'s the proper way to tie a square knot.',
             'title': 'How to Tie a Square Knot Properly',
-        }
+            'description': 'md5:dbe792e5f6f1489027027bf2eba188a3',
+            'timestamp': 1276081287,
+            'upload_date': '20100609',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
 
-        video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        self.report_extraction(video_id)
+        embed_code = self._search_regex(
-
+            r'<iframe[^>]+src="[^"]+\bembed_code=([^\b]+)\b',
-        video_url = self._search_regex(r'\'?file\'?: "(http://mobile-media\.howcast\.com/[0-9]+\.mp4)',
+            webpage, 'ooyala embed code')
-                                       webpage, 'video URL')
-
-        video_description = self._html_search_regex(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') name=\'description\'',
-                                                    webpage, 'description', fatal=False)
 
         return {
+            '_type': 'url_transparent',
+            'ie_key': 'Ooyala',
+            'url': 'ooyala:%s' % embed_code,
             'id': video_id,
-            'url': video_url,
+            'timestamp': parse_iso8601(self._html_search_meta(
-            'title': self._og_search_title(webpage),
+                'article:published_time', webpage, 'timestamp')),
-            'description': video_description,
-            'thumbnail': self._og_search_thumbnail(webpage),
         }

youtube_dl/extractor/ina.py

@@ -7,7 +7,7 @@ from .common import InfoExtractor
 
 
 class InaIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?ina\.fr/video/(?P<id>I?[A-Z0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?ina\.fr/video/(?P<id>I?[A-Z0-9]+)'
     _TEST = {
         'url': 'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
         'md5': 'a667021bf2b41f8dc6049479d9bb38a3',

youtube_dl/extractor/infoq.py

@@ -5,6 +5,7 @@ import base64
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
+    compat_urlparse,
 )
 
 
@@ -45,7 +46,7 @@ class InfoQIE(InfoExtractor):
         video_id, extension = video_filename.split('.')
 
         http_base = self._search_regex(
-            r'EXPRESSINSTALL_SWF\s*=\s*"(https?://[^/"]+/)', webpage,
+            r'EXPRESSINSTALL_SWF\s*=\s*[^"]*"((?:https?:)?//[^/"]+/)', webpage,
             'HTTP base URL')
 
         formats = [{
@@ -55,7 +56,7 @@ class InfoQIE(InfoExtractor):
             'play_path': playpath,
         }, {
             'format_id': 'http',
-            'url': http_base + real_id,
+            'url': compat_urlparse.urljoin(url, http_base) + real_id,
         }]
         self._sort_formats(formats)
 

youtube_dl/extractor/myvi.py

@@ -0,0 +1,60 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .vimple import SprutoBaseIE
+
+
+class MyviIE(SprutoBaseIE):
+    _VALID_URL = r'''(?x)
+                    https?://
+                        myvi\.(?:ru/player|tv)/
+                            (?:
+                                (?:
+                                    embed/html|
+                                    flash|
+                                    api/Video/Get
+                                )/|
+                                content/preloader\.swf\?.*\bid=
+                            )
+                            (?P<id>[\da-zA-Z_-]+)
+                    '''
+    _TESTS = [{
+        'url': 'http://myvi.ru/player/embed/html/oOy4euHA6LVwNNAjhD9_Jq5Ha2Qf0rtVMVFMAZav8wObeRTZaCATzucDQIDph8hQU0',
+        'md5': '571bbdfba9f9ed229dc6d34cc0f335bf',
+        'info_dict': {
+            'id': 'f16b2bbd-cde8-481c-a981-7cd48605df43',
+            'ext': 'mp4',
+            'title': 'хозяин жизни',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 25,
+        },
+    }, {
+        'url': 'http://myvi.ru/player/content/preloader.swf?id=oOy4euHA6LVwNNAjhD9_Jq5Ha2Qf0rtVMVFMAZav8wOYf1WFpPfc_bWTKGVf_Zafr0',
+        'only_matching': True,
+    }, {
+        'url': 'http://myvi.ru/player/api/Video/Get/oOy4euHA6LVwNNAjhD9_Jq5Ha2Qf0rtVMVFMAZav8wObeRTZaCATzucDQIDph8hQU0',
+        'only_matching': True,
+    }, {
+        'url': 'http://myvi.tv/embed/html/oTGTNWdyz4Zwy_u1nraolwZ1odenTd9WkTnRfIL9y8VOgHYqOHApE575x4_xxS9Vn0?ap=0',
+        'only_matching': True,
+    }, {
+        'url': 'http://myvi.ru/player/flash/ocp2qZrHI-eZnHKQBK4cZV60hslH8LALnk0uBfKsB-Q4WnY26SeGoYPi8HWHxu0O30',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def _extract_url(cls, webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//myvi\.(?:ru/player|tv)/(?:embed/html|flash)/[^"]+)\1', webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        spruto = self._download_json(
+            'http://myvi.ru/player/api/Video/Get/%s?sig' % video_id, video_id)['sprutoData']
+
+        return self._extract_spruto(spruto, video_id)

youtube_dl/extractor/npo.py

@@ -16,8 +16,24 @@ class NPOBaseIE(InfoExtractor):
         token_page = self._download_webpage(
             'http://ida.omroep.nl/npoplayer/i.js',
             video_id, note='Downloading token')
-        return self._search_regex(
+        token = self._search_regex(
             r'npoplayer\.token = "(.+?)"', token_page, 'token')
+        # Decryption algorithm extracted from http://npoplayer.omroep.nl/csjs/npoplayer-min.js
+        token_l = list(token)
+        first = second = None
+        for i in range(5, len(token_l) - 4):
+            if token_l[i].isdigit():
+                if first is None:
+                    first = i
+                elif second is None:
+                    second = i
+        if first is None or second is None:
+            first = 12
+            second = 13
+
+        token_l[first], token_l[second] = token_l[second], token_l[first]
+
+        return ''.join(token_l)
 
 
 class NPOIE(NPOBaseIE):
@@ -92,7 +108,7 @@ class NPOIE(NPOBaseIE):
 
     def _get_info(self, video_id):
         metadata = self._download_json(
-            'http://e.omroep.nl/metadata/aflevering/%s' % video_id,
+            'http://e.omroep.nl/metadata/%s' % video_id,
             video_id,
             # We have to remove the javascript callback
             transform_source=strip_jsonp,

youtube_dl/extractor/nrk.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 
 class NRKIE(InfoExtractor):
-    _VALID_URL = r'(?:nrk:|http://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
+    _VALID_URL = r'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
 
     _TESTS = [
         {
@@ -76,7 +76,7 @@ class NRKIE(InfoExtractor):
 
 
 class NRKPlaylistIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?nrk\.no/(?!video)(?:[^/]+/)+(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video)(?:[^/]+/)+(?P<id>[^/]+)'
 
     _TESTS = [{
         'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
@@ -116,11 +116,11 @@ class NRKPlaylistIE(InfoExtractor):
 
 
 class NRKTVIE(InfoExtractor):
-    _VALID_URL = r'(?P<baseurl>http://tv\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
+    _VALID_URL = r'(?P<baseurl>https?://tv\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
 
     _TESTS = [
         {
-            'url': 'http://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
+            'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
             'md5': 'adf2c5454fa2bf032f47a9f8fb351342',
             'info_dict': {
                 'id': 'MUHH48000314',
@@ -132,7 +132,7 @@ class NRKTVIE(InfoExtractor):
             },
         },
         {
-            'url': 'http://tv.nrk.no/program/mdfp15000514',
+            'url': 'https://tv.nrk.no/program/mdfp15000514',
             'md5': '383650ece2b25ecec996ad7b5bb2a384',
             'info_dict': {
                 'id': 'mdfp15000514',
@@ -145,7 +145,7 @@ class NRKTVIE(InfoExtractor):
         },
         {
             # single playlist video
-            'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
+            'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
             'md5': 'adbd1dbd813edaf532b0a253780719c2',
             'info_dict': {
                 'id': 'MSPO40010515-part2',
@@ -157,7 +157,7 @@ class NRKTVIE(InfoExtractor):
             'skip': 'Only works from Norway',
         },
         {
-            'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
+            'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
             'playlist': [
                 {
                     'md5': '9480285eff92d64f06e02a5367970a7a',

youtube_dl/extractor/pbs.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import re
@@ -35,6 +36,9 @@ class PBSIE(InfoExtractor):
                 'description': 'md5:ba0c207295339c8d6eced00b7c363c6a',
                 'duration': 3190,
             },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
@@ -46,6 +50,9 @@ class PBSIE(InfoExtractor):
                 'description': 'md5:f5bfbefadf421e8bb8647602011caf8e',
                 'duration': 5050,
             },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            }
         },
         {
             'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
@@ -68,7 +75,10 @@ class PBSIE(InfoExtractor):
                 'title': 'Dudamel Conducts Verdi Requiem at the Hollywood Bowl - Full',
                 'duration': 6559,
                 'thumbnail': 're:^https?://.*\.jpg$',
-            }
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/nova/earth/killer-typhoon.html',
@@ -82,7 +92,10 @@ class PBSIE(InfoExtractor):
                 'duration': 3172,
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'upload_date': '20140122',
-            }
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
@@ -90,6 +103,21 @@ class PBSIE(InfoExtractor):
                 'id': 'united-states-of-secrets',
             },
             'playlist_count': 2,
+        },
+        {
+            'url': 'http://www.pbs.org/wgbh/americanexperience/films/death/player/',
+            'info_dict': {
+                'id': '2280706814',
+                'display_id': 'player',
+                'ext': 'mp4',
+                'title': 'Death and the Civil War',
+                'description': 'American Experience, TV’s most-watched history series, brings to life the compelling stories from our past that inform our understanding of the world today.',
+                'duration': 6705,
+                'thumbnail': 're:^https?://.*\.jpg$',
+            },
+            'params': {
+                'skip_download': True,  # requires ffmpeg
+            },
         }
     ]
 
@@ -123,7 +151,7 @@ class PBSIE(InfoExtractor):
                 return media_id, presumptive_id, upload_date
 
             url = self._search_regex(
-                r'<iframe\s+(?:class|id)=["\']partnerPlayer["\'].*?\s+src=["\'](.*?)["\']>',
+                r'<iframe\s+[^>]*\s+src=["\']([^\'"]+partnerplayer[^\'"]+)["\']',
                 webpage, 'player URL')
             mobj = re.match(self._VALID_URL, url)
 
@@ -196,6 +224,14 @@ class PBSIE(InfoExtractor):
             rating_str = rating_str.rpartition('-')[2]
         age_limit = US_RATINGS.get(rating_str)
 
+        subtitles = {}
+        closed_captions_url = info.get('closed_captions_url')
+        if closed_captions_url:
+            subtitles['en'] = [{
+                'ext': 'ttml',
+                'url': closed_captions_url,
+            }]
+
         return {
             'id': video_id,
             'display_id': display_id,
@@ -206,4 +242,5 @@ class PBSIE(InfoExtractor):
             'age_limit': age_limit,
             'upload_date': upload_date,
             'formats': formats,
+            'subtitles': subtitles,
         }

youtube_dl/extractor/qqmusic.py

@@ -9,6 +9,7 @@ from .common import InfoExtractor
 from ..utils import (
     strip_jsonp,
     unescapeHTML,
+    clean_html,
 )
 from ..compat import compat_urllib_request
 
@@ -26,6 +27,20 @@ class QQMusicIE(InfoExtractor):
             'upload_date': '20141227',
             'creator': '林俊杰',
             'description': 'md5:d327722d0361576fde558f1ac68a7065',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        }
+    }, {
+        'note': 'There is no mp3-320 version of this song.',
+        'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV',
+        'md5': 'fa3926f0c585cda0af8fa4f796482e3e',
+        'info_dict': {
+            'id': '004MsGEo3DdNxV',
+            'ext': 'mp3',
+            'title': '如果',
+            'upload_date': '20050626',
+            'creator': '李季美',
+            'description': 'md5:46857d5ed62bc4ba84607a805dccf437',
+            'thumbnail': 're:^https?://.*\.jpg$',
         }
     }]
 
@@ -68,6 +83,14 @@ class QQMusicIE(InfoExtractor):
         if lrc_content:
             lrc_content = lrc_content.replace('\\n', '\n')
 
+        thumbnail_url = None
+        albummid = self._search_regex(
+            [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'],
+            detail_info_page, 'album mid', default=None)
+        if albummid:
+            thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \
+                            % (albummid[-2:-1], albummid[-1], albummid)
+
         guid = self.m_r_get_ruin()
 
         vkey = self._download_json(
@@ -85,6 +108,7 @@ class QQMusicIE(InfoExtractor):
                 'preference': details['preference'],
                 'abr': details.get('abr'),
             })
+        self._check_formats(formats, mid)
         self._sort_formats(formats)
 
         return {
@@ -94,6 +118,7 @@ class QQMusicIE(InfoExtractor):
             'upload_date': publish_time,
             'creator': singer,
             'description': lrc_content,
+            'thumbnail': thumbnail_url,
         }
 
 
@@ -163,31 +188,40 @@ class QQMusicAlbumIE(QQPlaylistBaseIE):
     IE_NAME = 'qqmusic:album'
     _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
 
-    _TEST = {
+    _TESTS = [{
-        'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
+        'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1',
         'info_dict': {
             'id': '000gXCTb2AhRR1',
             'title': '我们都是这样长大的',
-            'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
+            'description': 'md5:179c5dce203a5931970d306aa9607ea6',
         },
         'playlist_count': 4,
-    }
+    }, {
+        'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3',
+        'info_dict': {
+            'id': '002Y5a3b3AlCu3',
+            'title': '그리고...',
+            'description': 'md5:a48823755615508a95080e81b51ba729',
+        },
+        'playlist_count': 8,
+    }]
 
     def _real_extract(self, url):
         mid = self._match_id(url)
 
-        album_page = self._download_webpage(
+        album = self._download_json(
-            self.qq_static_url('album', mid), mid, 'Download album page')
+            'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid,
+            mid, 'Download album page')['data']
 
-        entries = self.get_entries_from_page(album_page)
+        entries = [
-
+            self.url_result(
-        album_name = self._html_search_regex(
+                'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
-            r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
+            ) for song in album['list']
-            default=None)
+        ]
-
+        album_name = album.get('name')
-        album_detail = self._html_search_regex(
+        album_detail = album.get('desc')
-            r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
+        if album_detail is not None:
-            album_page, 'album details', default=None)
+            album_detail = album_detail.strip()
 
         return self.playlist_result(entries, mid, album_name, album_detail)
 
@@ -243,3 +277,36 @@ class QQMusicToplistIE(QQPlaylistBaseIE):
         list_name = topinfo.get('ListName')
         list_description = topinfo.get('info')
         return self.playlist_result(entries, list_id, list_name, list_description)
+
+
+class QQMusicPlaylistIE(QQPlaylistBaseIE):
+    IE_NAME = 'qqmusic:playlist'
+    _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
+
+    _TEST = {
+        'url': 'http://y.qq.com/#type=taoge&id=3462654915',
+        'info_dict': {
+            'id': '3462654915',
+            'title': '韩国5月新歌精选下旬',
+            'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
+        },
+        'playlist_count': 40,
+    }
+
+    def _real_extract(self, url):
+        list_id = self._match_id(url)
+
+        list_json = self._download_json(
+            'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg?type=1&json=1&utf8=1&onlysong=0&disstid=%s'
+            % list_id, list_id, 'Download list page',
+            transform_source=strip_jsonp)['cdlist'][0]
+
+        entries = [
+            self.url_result(
+                'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
+            ) for song in list_json['songlist']
+        ]
+
+        list_name = list_json.get('dissname')
+        list_description = clean_html(unescapeHTML(list_json.get('desc')))
+        return self.playlist_result(entries, list_id, list_name, list_description)

youtube_dl/extractor/rtlnl.py

@@ -43,6 +43,10 @@ class RtlNlIE(InfoExtractor):
             'upload_date': '20150215',
             'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.',
         }
+    }, {
+        # encrypted m3u8 streams, georestricted
+        'url': 'http://www.rtlxl.nl/#!/afl-2-257632/52a74543-c504-4cde-8aa8-ec66fe8d68a7',
+        'only_matching': True,
     }, {
         'url': 'http://www.rtl.nl/system/videoplayer/derden/embed.html#!/uuid=bb0353b0-d6a4-1dad-90e9-18fe75b8d1f0',
         'only_matching': True,
@@ -51,7 +55,7 @@ class RtlNlIE(InfoExtractor):
     def _real_extract(self, url):
         uuid = self._match_id(url)
         info = self._download_json(
-            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=flash/' % uuid,
+            'http://www.rtl.nl/system/s4m/vfd/version=2/uuid=%s/fmt=adaptive/' % uuid,
             uuid)
 
         material = info['material'][0]
@@ -59,9 +63,14 @@ class RtlNlIE(InfoExtractor):
         subtitle = material['title'] or info['episodes'][0]['name']
         description = material.get('synopsis') or info['episodes'][0]['synopsis']
 
+        meta = info.get('meta', {})
+
         # Use unencrypted m3u8 streams (See https://github.com/rg3/youtube-dl/issues/4118)
-        videopath = material['videopath'].replace('.f4m', '.m3u8')
+        # NB: nowadays, recent ffmpeg and avconv can handle these encrypted streams, so
-        m3u8_url = 'http://manifest.us.rtl.nl' + videopath
+        # this adaptive -> flash workaround is not required in general, but it also
+        # allows bypassing georestriction therefore is retained for now.
+        videopath = material['videopath'].replace('/adaptive/', '/flash/')
+        m3u8_url = meta.get('videohost', 'http://manifest.us.rtl.nl') + videopath
 
         formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4')
 
@@ -82,7 +91,7 @@ class RtlNlIE(InfoExtractor):
         self._sort_formats(formats)
 
         thumbnails = []
-        meta = info.get('meta', {})
+
         for p in ('poster_base_url', '"thumb_base_url"'):
             if not meta.get(p):
                 continue

youtube_dl/extractor/spiegeltv.py

@@ -77,11 +77,13 @@ class SpiegeltvIE(InfoExtractor):
                     'rtmp_live': True,
                 })
             elif determine_ext(endpoint) == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
+                m3u8_formats = self._extract_m3u8_formats(
                     endpoint.replace('[video]', play_path),
                     video_id, 'm4v',
                     preference=1,  # Prefer hls since it allows to workaround georestriction
-                    m3u8_id='hls'))
+                    m3u8_id='hls', fatal=False)
+                if m3u8_formats is not False:
+                    formats.extend(m3u8_formats)
             else:
                 formats.append({
                     'url': endpoint,

youtube_dl/extractor/thisamericanlife.py

@@ -0,0 +1,40 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class ThisAmericanLifeIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?thisamericanlife\.org/(?:radio-archives/episode/|play_full\.php\?play=)(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://www.thisamericanlife.org/radio-archives/episode/487/harper-high-school-part-one',
+        'md5': '8f7d2da8926298fdfca2ee37764c11ce',
+        'info_dict': {
+            'id': '487',
+            'ext': 'm4a',
+            'title': '487: Harper High School, Part One',
+            'description': 'md5:ee40bdf3fb96174a9027f76dbecea655',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
+    }, {
+        'url': 'http://www.thisamericanlife.org/play_full.php?play=487',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://www.thisamericanlife.org/radio-archives/episode/%s' % video_id, video_id)
+
+        return {
+            'id': video_id,
+            'url': 'http://stream.thisamericanlife.org/{0}/stream/{0}_64k.m3u8'.format(video_id),
+            'protocol': 'm3u8_native',
+            'ext': 'm4a',
+            'acodec': 'aac',
+            'vcodec': 'none',
+            'abr': 64,
+            'title': self._html_search_meta(r'twitter:title', webpage, 'title', fatal=True),
+            'description': self._html_search_meta(r'description', webpage, 'description'),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }

youtube_dl/extractor/vimple.py

@@ -4,7 +4,29 @@ from .common import InfoExtractor
 from ..utils import int_or_none
 
 
-class VimpleIE(InfoExtractor):
+class SprutoBaseIE(InfoExtractor):
+    def _extract_spruto(self, spruto, video_id):
+        playlist = spruto['playlist'][0]
+        title = playlist['title']
+        video_id = playlist.get('videoId') or video_id
+        thumbnail = playlist.get('posterUrl') or playlist.get('thumbnailUrl')
+        duration = int_or_none(playlist.get('duration'))
+
+        formats = [{
+            'url': f['url'],
+        } for f in playlist['video']]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }
+
+
+class VimpleIE(SprutoBaseIE):
     IE_DESC = 'Vimple - one-click video hosting'
     _VALID_URL = r'https?://(?:player\.vimple\.ru/iframe|vimple\.ru)/(?P<id>[\da-f-]{32,36})'
     _TESTS = [
@@ -30,25 +52,9 @@ class VimpleIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://player.vimple.ru/iframe/%s' % video_id, video_id)
 
-        playlist = self._parse_json(
+        spruto = self._parse_json(
             self._search_regex(
                 r'sprutoData\s*:\s*({.+?}),\r\n', webpage, 'spruto data'),
-            video_id)['playlist'][0]
+            video_id)
 
-        title = playlist['title']
+        return self._extract_spruto(spruto, video_id)
-        video_id = playlist.get('videoId') or video_id
-        thumbnail = playlist.get('posterUrl') or playlist.get('thumbnailUrl')
-        duration = int_or_none(playlist.get('duration'))
-
-        formats = [{
-            'url': f['url'],
-        } for f in playlist['video']]
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-        }

youtube_dl/extractor/vk.py

@@ -21,7 +21,17 @@ from ..utils import (
 
 class VKIE(InfoExtractor):
     IE_NAME = 'vk.com'
-    _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|(?:.+?\?.*?z=)?video(?P<videoid>[^s].*?)(?:\?|%2F|$))'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:
+                            (?:m\.)?vk\.com/video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|
+                            (?:
+                                (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video|
+                                (?:www\.)?biqle\.ru/watch/
+                            )
+                            (?P<videoid>[^s].*?)(?:\?|%2F|$)
+                        )
+                    '''
     _NETRC_MACHINE = 'vk'
 
     _TESTS = [
@@ -109,11 +119,31 @@ class VKIE(InfoExtractor):
             },
             'skip': 'Only works from Russia',
         },
+        {
+            # youtube embed
+            'url': 'https://vk.com/video276849682_170681728',
+            'info_dict': {
+                'id': 'V3K4mi0SYkc',
+                'ext': 'mp4',
+                'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
+                'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
+                'duration': 179,
+                'upload_date': '20130116',
+                'uploader': "Children's Joy Foundation",
+                'uploader_id': 'thecjf',
+                'view_count': int,
+            },
+        },
         {
             # removed video, just testing that we match the pattern
             'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
             'only_matching': True,
         },
+        {
+            # vk wrapper
+            'url': 'http://www.biqle.ru/watch/847655_160197695',
+            'only_matching': True,
+        }
     ]
 
     def _login(self):
@@ -153,9 +183,14 @@ class VKIE(InfoExtractor):
         if not video_id:
             video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
 
-        info_url = 'http://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
+        info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
         info_page = self._download_webpage(info_url, video_id)
 
+        if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
+            raise ExtractorError(
+                'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
+                expected=True)
+
         ERRORS = {
             r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
             'Video %s has been removed from public access due to rightholder complaint.',
@@ -175,10 +210,11 @@ class VKIE(InfoExtractor):
             if re.search(error_re, info_page):
                 raise ExtractorError(error_msg % video_id, expected=True)
 
-        m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page)
+        youtube_url = self._search_regex(
-        if m_yt is not None:
+            r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
-            self.to_screen('Youtube video detected')
+            info_page, 'youtube iframe', default=None)
-            return self.url_result(m_yt.group(1), 'Youtube')
+        if youtube_url:
+            return self.url_result(youtube_url, 'Youtube')
 
         m_rutube = re.search(
             r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page)

youtube_dl/extractor/yinyuetai.py

@@ -0,0 +1,55 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import ExtractorError
+
+
+class YinYueTaiIE(InfoExtractor):
+    IE_NAME = 'yinyuetai:video'
+    _VALID_URL = r'https?://v\.yinyuetai\.com/video(?:/h5)?/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://v.yinyuetai.com/video/2322376',
+        'md5': '6e3abe28d38e3a54b591f9f040595ce0',
+        'info_dict': {
+            'id': '2322376',
+            'ext': 'mp4',
+            'title': '少女时代_PARTY_Music Video Teaser',
+            'creator': '少女时代',
+            'duration': 25,
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
+    }, {
+        'url': 'http://v.yinyuetai.com/video/h5/2322376',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        info = self._download_json(
+            'http://ext.yinyuetai.com/main/get-h-mv-info?json=true&videoId=%s' % video_id, video_id,
+            'Downloading mv info')['videoInfo']['coreVideoInfo']
+
+        if info['error']:
+            raise ExtractorError(info['errorMsg'], expected=True)
+
+        formats = [{
+            'url': format_info['videoUrl'],
+            'format_id': format_info['qualityLevel'],
+            'format': format_info.get('qualityLevelName'),
+            'filesize': format_info.get('fileSize'),
+            # though URLs ends with .flv, the downloaded files are in fact mp4
+            'ext': 'mp4',
+            'tbr': format_info.get('bitrate'),
+        } for format_info in info['videoUrlModels']]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': info['videoName'],
+            'thumbnail': info.get('bigHeadImage'),
+            'creator': info.get('artistNames'),
+            'duration': info.get('duration'),
+            'formats': formats,
+        }

youtube_dl/extractor/youtube.py

@@ -520,6 +520,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'skip_download': 'requires avconv',
             }
         },
+        # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
+        {
+            'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
+            'info_dict': {
+                'id': 'FIl7x6_3R5Y',
+                'ext': 'mp4',
+                'title': 'md5:7b81415841e02ecd4313668cde88737a',
+                'description': 'md5:116377fd2963b81ec4ce64b542173306',
+                'upload_date': '20150625',
+                'uploader_id': 'dorappi2000',
+                'uploader': 'dorappi2000',
+                'formats': 'mincount:33',
+            },
+        }
     ]
 
     def __init__(self, *args, **kwargs):
@@ -784,7 +798,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         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):
+            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)
@@ -793,7 +807,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         dash_doc = self._download_xml(
             dash_manifest_url, video_id,
             note='Downloading DASH manifest',
-            errnote='Could not download 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'):
@@ -826,6 +844,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     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' and 'vcodec' not in full_info:
+                                full_info['vcodec'] = codecs
+                            elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
+                                full_info['acodec'] = codecs
                         formats.append(full_info)
                     else:
                         existing_format.update(f)
@@ -855,6 +879,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         else:
             player_url = None
 
+        dash_mpds = []
+
+        def add_dash_mpd(video_info):
+            dash_mpd = video_info.get('dashmpd')
+            if dash_mpd and dash_mpd[0] not in dash_mpds:
+                dash_mpds.append(dash_mpd[0])
+
         # Get video info
         embed_webpage = None
         if re.search(r'player-age-gate-content">', video_webpage) is not None:
@@ -875,24 +906,29 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 note='Refetching age-gated info webpage',
                 errnote='unable to download video info webpage')
             video_info = compat_parse_qs(video_info_webpage)
+            add_dash_mpd(video_info)
         else:
             age_gate = False
-            try:
+            video_info = None
             # Try looking directly into the video webpage
             mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
-                if not mobj:
+            if mobj:
-                    raise ValueError('Could not find ytplayer.config')  # caught below
                 json_code = uppercase_escape(mobj.group(1))
                 ytplayer_config = json.loads(json_code)
                 args = ytplayer_config['args']
+                if args.get('url_encoded_fmt_stream_map'):
                     # Convert to the same format returned by compat_parse_qs
                     video_info = dict((k, [v]) for k, v in args.items())
-                if not args.get('url_encoded_fmt_stream_map'):
+                    add_dash_mpd(video_info)
-                    raise ValueError('No stream_map present')  # caught below
+            if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
-            except ValueError:
+                # We also try looking in get_video_info since it may contain different dashmpd
-                # We fallback to the get_video_info pages (used by the embed page)
+                # URL that points to a DASH manifest with possibly different itag set (some itags
+                # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
+                # manifest pointed by get_video_info's dashmpd).
+                # The general idea is to take a union of itags of both DASH manifests (for example
+                # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
                 self.report_video_info_webpage_download(video_id)
-                for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
+                for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
                     video_info_url = (
                         '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
                         % (proto, video_id, el_type))
@@ -900,8 +936,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                         video_info_url,
                         video_id, note=False,
                         errnote='unable to download video info webpage')
-                    video_info = compat_parse_qs(video_info_webpage)
+                    get_video_info = compat_parse_qs(video_info_webpage)
-                    if 'token' in video_info:
+                    add_dash_mpd(get_video_info)
+                    if not video_info:
+                        video_info = get_video_info
+                    if 'token' in get_video_info:
                         break
         if 'token' not in video_info:
             if 'reason' in video_info:
@@ -964,13 +1003,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
 
         # upload date
-        upload_date = None
+        upload_date = self._html_search_meta(
-        mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
+            'datePublished', video_webpage, 'upload date', default=None)
-        if mobj is None:
+        if not upload_date:
-            mobj = re.search(
+            upload_date = self._search_regex(
-                r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
+                [r'(?s)id="eow-date.*?>(.*?)</span>',
-                video_webpage)
+                 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
-        if mobj is not None:
+                video_webpage, 'upload date', default=None)
+            if upload_date:
                 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
         upload_date = unified_strdate(upload_date)
 
@@ -1125,24 +1165,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         # Look for the DASH manifest
         if self._downloader.params.get('youtube_include_dash_manifest', True):
-            dash_mpd = video_info.get('dashmpd')
+            dash_mpd_fatal = True
-            if dash_mpd:
+            for dash_manifest_url in dash_mpds:
-                dash_manifest_url = dash_mpd[0]
+                dash_formats = {}
                 try:
-                    dash_formats = self._parse_dash_manifest(
+                    for df in self._parse_dash_manifest(
-                        video_id, dash_manifest_url, player_url, age_gate)
+                            video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
+                        # 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
+                        # Additional DASH manifests may end up in HTTP Error 403 therefore
+                        # allow them to fail without bug report message if we already have
+                        # some DASH manifest succeeded. This is temporary workaround to reduce
+                        # burst of bug reports until we figure out the reason and whether it
+                        # can be fixed at all.
+                        dash_mpd_fatal = False
                 except (ExtractorError, KeyError) as e:
                     self.report_warning(
                         'Skipping DASH manifest: %r' % e, video_id)
-                else:
+                if dash_formats:
                     # Remove the formats we found through non-DASH, they
                     # contain less info and it can be wrong, because we use
                     # fixed values (for example the resolution). See
                     # https://github.com/rg3/youtube-dl/issues/5774 for an
                     # example.
-                    dash_keys = set(df['format_id'] for df in dash_formats)
+                    formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
-                    formats = [f for f in formats if f['format_id'] not in dash_keys]
+                    formats.extend(dash_formats.values())
-                    formats.extend(dash_formats)
 
         # Check for malformed aspect ratio
         stretched_m = re.search(

youtube_dl/options.py

@@ -346,7 +346,7 @@ def parseOpts(overrideArguments=None):
     video_format.add_option(
         '--youtube-skip-dash-manifest',
         action='store_false', dest='youtube_include_dash_manifest',
-        help='Do not download the DASH manifest on YouTube videos')
+        help='Do not download the DASH manifests and related data on YouTube videos')
     video_format.add_option(
         '--merge-output-format',
         action='store', dest='merge_output_format', metavar='FORMAT', default=None,

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.06.25'
+__version__ = '2015.07.07'