release 2017.06.23

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.06.18*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.06.23*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.06.18**
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2017.06.23**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2017.06.18
+[debug] youtube-dl version 2017.06.23
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

ChangeLog

@@ -1,3 +1,24 @@
+version 2017.06.23
+
+Core
+* [adobepass] Fix extraction on older python 2.6
+
+Extractors
+* [youtube] Adapt to new automatic captions rendition (#13467)
+* [hgtv.com:show] Relax video config regular expression (#13279, #13461)
+* [drtuber] Fix formats extraction (#12058)
+* [youporn] Fix upload date extraction
+* [youporn] Improve formats extraction
+* [youporn] Fix title extraction (#13456)
+* [googledrive] Fix formats sorting (#13443)
+* [watchindianporn] Fix extraction (#13411, #13415)
++ [vimeo] Add fallback mp4 extension for original format
++ [ruv] Add support for ruv.is (#13396)
+* [viu] Fix extraction on older python 2.6
+* [pandora.tv] Fix upload_date extraction (#12846)
++ [asiancrush] Add support for asiancrush.com (#13420)
+
+
 version 2017.06.18
 
 Core

docs/supportedsites.md

@@ -67,6 +67,8 @@
  - **arte.tv:info**
  - **arte.tv:magazine**
  - **arte.tv:playlist**
+ - **AsianCrush**
+ - **AsianCrushPlaylist**
  - **AtresPlayer**
  - **ATTTechChannel**
  - **ATVAt**
@@ -686,6 +688,7 @@
  - **rutube:person**: Rutube person videos
  - **RUTV**: RUTV.RU
  - **Ruutu**
+ - **Ruv**
  - **safari**: safaribooksonline.com online video
  - **safari:api**
  - **safari:course**: safaribooksonline.com online courses

youtube_dl/extractor/abcotvs.py

@@ -22,7 +22,7 @@ class ABCOTVSIE(InfoExtractor):
                 'display_id': 'east-bay-museum-celebrates-vintage-synthesizers',
                 'ext': 'mp4',
                 'title': 'East Bay museum celebrates vintage synthesizers',
-                'description': 'md5:a4f10fb2f2a02565c1749d4adbab4b10',
+                'description': 'md5:24ed2bd527096ec2a5c67b9d5a9005f3',
                 'thumbnail': r're:^https?://.*\.jpg$',
                 'timestamp': 1421123075,
                 'upload_date': '20150113',

youtube_dl/extractor/adobepass.py

@@ -6,7 +6,10 @@ import time
 import xml.etree.ElementTree as etree
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_kwargs,
+    compat_urlparse,
+)
 from ..utils import (
     unescapeHTML,
     urlencode_postdata,
@@ -1317,7 +1320,8 @@ class AdobePassIE(InfoExtractor):
         headers = kwargs.get('headers', {})
         headers.update(self.geo_verification_headers())
         kwargs['headers'] = headers
-        return super(AdobePassIE, self)._download_webpage_handle(*args, **kwargs)
+        return super(AdobePassIE, self)._download_webpage_handle(
+            *args, **compat_kwargs(kwargs))
 
     @staticmethod
     def _get_mvpd_resource(provider_id, title, guid, rating):

youtube_dl/extractor/asiancrush.py

@@ -0,0 +1,93 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from .kaltura import KalturaIE
+from ..utils import (
+    extract_attributes,
+    remove_end,
+    urlencode_postdata,
+)
+
+
+class AsianCrushIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?asiancrush\.com/video/(?:[^/]+/)?0+(?P<id>\d+)v\b'
+    _TESTS = [{
+        'url': 'https://www.asiancrush.com/video/012869v/women-who-flirt/',
+        'md5': 'c3b740e48d0ba002a42c0b72857beae6',
+        'info_dict': {
+            'id': '1_y4tmjm5r',
+            'ext': 'mp4',
+            'title': 'Women Who Flirt',
+            'description': 'md5:3db14e9186197857e7063522cb89a805',
+            'timestamp': 1496936429,
+            'upload_date': '20170608',
+            'uploader_id': 'craig@crifkin.com',
+        },
+    }, {
+        'url': 'https://www.asiancrush.com/video/she-was-pretty/011886v-pretty-episode-3/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        data = self._download_json(
+            'https://www.asiancrush.com/wp-admin/admin-ajax.php', video_id,
+            data=urlencode_postdata({
+                'postid': video_id,
+                'action': 'get_channel_kaltura_vars',
+            }))
+
+        entry_id = data['entry_id']
+
+        return self.url_result(
+            'kaltura:%s:%s' % (data['partner_id'], entry_id),
+            ie=KalturaIE.ie_key(), video_id=entry_id,
+            video_title=data.get('vid_label'))
+
+
+class AsianCrushPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?asiancrush\.com/series/0+(?P<id>\d+)s\b'
+    _TEST = {
+        'url': 'https://www.asiancrush.com/series/012481s/scholar-walks-night/',
+        'info_dict': {
+            'id': '12481',
+            'title': 'Scholar Who Walks the Night',
+            'description': 'md5:7addd7c5132a09fd4741152d96cce886',
+        },
+        'playlist_count': 20,
+    }
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = []
+
+        for mobj in re.finditer(
+                r'<a[^>]+href=(["\'])(?P<url>%s.*?)\1[^>]*>' % AsianCrushIE._VALID_URL,
+                webpage):
+            attrs = extract_attributes(mobj.group(0))
+            if attrs.get('class') == 'clearfix':
+                entries.append(self.url_result(
+                    mobj.group('url'), ie=AsianCrushIE.ie_key()))
+
+        title = remove_end(
+            self._html_search_regex(
+                r'(?s)<h1\b[^>]\bid=["\']movieTitle[^>]+>(.+?)</h1>', webpage,
+                'title', default=None) or self._og_search_title(
+                webpage, default=None) or self._html_search_meta(
+                'twitter:title', webpage, 'title',
+                default=None) or self._search_regex(
+                r'<title>([^<]+)</title>', webpage, 'title', fatal=False),
+            ' | AsianCrush')
+
+        description = self._og_search_description(
+            webpage, default=None) or self._html_search_meta(
+            'twitter:description', webpage, 'description', fatal=False)
+
+        return self.playlist_result(entries, playlist_id, title, description)

youtube_dl/extractor/drtuber.py

@@ -44,8 +44,23 @@ class DrTuberIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://www.drtuber.com/video/%s' % video_id, display_id)
 
-        video_url = self._html_search_regex(
+        video_data = self._download_json(
-            r'<source src="([^"]+)"', webpage, 'video URL')
+            'http://www.drtuber.com/player_config_json/', video_id, query={
+                'vid': video_id,
+                'embed': 0,
+                'aid': 0,
+                'domain_id': 0,
+            })
+
+        formats = []
+        for format_id, video_url in video_data['files'].items():
+            if video_url:
+                formats.append({
+                    'format_id': format_id,
+                    'quality': 2 if format_id == 'hq' else 1,
+                    'url': video_url
+                })
+        self._sort_formats(formats)
 
         title = self._html_search_regex(
             (r'class="title_watch"[^>]*><(?:p|h\d+)[^>]*>([^<]+)<',
@@ -75,7 +90,7 @@ class DrTuberIE(InfoExtractor):
         return {
             'id': video_id,
             'display_id': display_id,
-            'url': video_url,
+            'formats': formats,
             'title': title,
             'thumbnail': thumbnail,
             'like_count': like_count,

youtube_dl/extractor/extractors.py

@@ -71,6 +71,10 @@ from .arte import (
     TheOperaPlatformIE,
     ArteTVPlaylistIE,
 )
+from .asiancrush import (
+    AsianCrushIE,
+    AsianCrushPlaylistIE,
+)
 from .atresplayer import AtresPlayerIE
 from .atttechchannel import ATTTechChannelIE
 from .atvat import ATVAtIE
@@ -871,6 +875,7 @@ from .rutube import (
 )
 from .rutv import RUTVIE
 from .ruutu import RuutuIE
+from .ruv import RuvIE
 from .sandia import SandiaIE
 from .safari import (
     SafariIE,

youtube_dl/extractor/googledrive.py

@@ -69,19 +69,32 @@ class GoogleDriveIE(InfoExtractor):
             r'"fmt_stream_map"\s*,\s*"([^"]+)', webpage, 'fmt stream map').split(',')
         fmt_list = self._search_regex(r'"fmt_list"\s*,\s*"([^"]+)', webpage, 'fmt_list').split(',')
 
+        resolutions = {}
+        for fmt in fmt_list:
+            mobj = re.search(
+                r'^(?P<format_id>\d+)/(?P<width>\d+)[xX](?P<height>\d+)', fmt)
+            if mobj:
+                resolutions[mobj.group('format_id')] = (
+                    int(mobj.group('width')), int(mobj.group('height')))
+
         formats = []
-        for fmt, fmt_stream in zip(fmt_list, fmt_stream_map):
+        for fmt_stream in fmt_stream_map:
-            fmt_id, fmt_url = fmt_stream.split('|')
+            fmt_stream_split = fmt_stream.split('|')
-            resolution = fmt.split('/')[1]
+            if len(fmt_stream_split) < 2:
-            width, height = resolution.split('x')
+                continue
-            formats.append({
+            format_id, format_url = fmt_stream_split[:2]
-                'url': lowercase_escape(fmt_url),
+            f = {
-                'format_id': fmt_id,
+                'url': lowercase_escape(format_url),
-                'resolution': resolution,
+                'format_id': format_id,
-                'width': int_or_none(width),
+                'ext': self._FORMATS_EXT[format_id],
-                'height': int_or_none(height),
+            }
-                'ext': self._FORMATS_EXT[fmt_id],
+            resolution = resolutions.get(format_id)
+            if resolution:
+                f.update({
+                    'width': resolution[0],
+                    'height': resolution[0],
                 })
+            formats.append(f)
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/hgtv.py

@@ -7,14 +7,19 @@ from .common import InfoExtractor
 class HGTVComShowIE(InfoExtractor):
     IE_NAME = 'hgtv.com:show'
     _VALID_URL = r'https?://(?:www\.)?hgtv\.com/shows/[^/]+/(?P<id>[^/?#&]+)'
-    _TEST = {
+    _TESTS = [{
-        'url': 'http://www.hgtv.com/shows/flip-or-flop/flip-or-flop-full-episodes-videos',
+        # data-module="video"
+        'url': 'http://www.hgtv.com/shows/flip-or-flop/flip-or-flop-full-episodes-season-4-videos',
         'info_dict': {
-            'id': 'flip-or-flop-full-episodes-videos',
+            'id': 'flip-or-flop-full-episodes-season-4-videos',
             'title': 'Flip or Flop Full Episodes',
         },
         'playlist_mincount': 15,
-    }
+    }, {
+        # data-deferred-module="video"
+        'url': 'http://www.hgtv.com/shows/good-bones/episodes/an-old-victorian-house-gets-a-new-facelift',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
@@ -23,7 +28,7 @@ class HGTVComShowIE(InfoExtractor):
 
         config = self._parse_json(
             self._search_regex(
-                r'(?s)data-module=["\']video["\'][^>]*>.*?<script[^>]+type=["\']text/x-config["\'][^>]*>(.+?)</script',
+                r'(?s)data-(?:deferred)?-module=["\']video["\'][^>]*>.*?<script[^>]+type=["\']text/x-config["\'][^>]*>(.+?)</script',
                 webpage, 'video config'),
             display_id)['channels'][0]
 

youtube_dl/extractor/pandoratv.py

@@ -19,7 +19,7 @@ class PandoraTVIE(InfoExtractor):
     IE_NAME = 'pandora.tv'
     IE_DESC = '판도라TV'
     _VALID_URL = r'https?://(?:.+?\.)?channel\.pandora\.tv/channel/video\.ptv\?'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://jp.channel.pandora.tv/channel/video.ptv?c1=&prgid=53294230&ch_userid=mikakim&ref=main&lot=cate_01_2',
         'info_dict': {
             'id': '53294230',
@@ -34,7 +34,26 @@ class PandoraTVIE(InfoExtractor):
             'view_count': int,
             'like_count': int,
         }
-    }
+    }, {
+        'url': 'http://channel.pandora.tv/channel/video.ptv?ch_userid=gogoucc&prgid=54721744',
+        'info_dict': {
+            'id': '54721744',
+            'ext': 'flv',
+            'title': '[HD] JAPAN COUNTDOWN 170423',
+            'description': '[HD] JAPAN COUNTDOWN 170423',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1704.9,
+            'upload_date': '20170423',
+            'uploader': 'GOGO_UCC',
+            'uploader_id': 'gogoucc',
+            'view_count': int,
+            'like_count': int,
+        },
+        'params': {
+            # Test metadata only
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
@@ -86,7 +105,7 @@ class PandoraTVIE(InfoExtractor):
             'description': info.get('body'),
             'thumbnail': info.get('thumbnail') or info.get('poster'),
             'duration': float_or_none(info.get('runtime'), 1000) or parse_duration(info.get('time')),
-            'upload_date': info['fid'][:8] if isinstance(info.get('fid'), compat_str) else None,
+            'upload_date': info['fid'].split('/')[-1][:8] if isinstance(info.get('fid'), compat_str) else None,
             'uploader': info.get('nickname'),
             'uploader_id': info.get('upload_userid'),
             'view_count': str_to_int(info.get('hit')),

youtube_dl/extractor/ruv.py

@@ -0,0 +1,101 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    unified_timestamp,
+)
+
+
+class RuvIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ruv\.is/(?:sarpurinn/[^/]+|node)/(?P<id>[^/]+(?:/\d+)?)'
+    _TESTS = [{
+        # m3u8
+        'url': 'http://ruv.is/sarpurinn/ruv-aukaras/fh-valur/20170516',
+        'md5': '66347652f4e13e71936817102acc1724',
+        'info_dict': {
+            'id': '1144499',
+            'display_id': 'fh-valur/20170516',
+            'ext': 'mp4',
+            'title': 'FH - Valur',
+            'description': 'Bein útsending frá 3. leik FH og Vals í úrslitum Olísdeildar karla í handbolta.',
+            'timestamp': 1494963600,
+            'upload_date': '20170516',
+        },
+    }, {
+        # mp3
+        'url': 'http://ruv.is/sarpurinn/ras-2/morgunutvarpid/20170619',
+        'md5': '395ea250c8a13e5fdb39d4670ef85378',
+        'info_dict': {
+            'id': '1153630',
+            'display_id': 'morgunutvarpid/20170619',
+            'ext': 'mp3',
+            'title': 'Morgunútvarpið',
+            'description': 'md5:a4cf1202c0a1645ca096b06525915418',
+            'timestamp': 1497855000,
+            'upload_date': '20170619',
+        },
+    }, {
+        'url': 'http://ruv.is/sarpurinn/ruv/frettir/20170614',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.ruv.is/node/1151854',
+        'only_matching': True,
+    }, {
+        'url': 'http://ruv.is/sarpurinn/klippa/secret-soltice-hefst-a-morgun',
+        'only_matching': True,
+    }, {
+        'url': 'http://ruv.is/sarpurinn/ras-1/morgunvaktin/20170619',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._og_search_title(webpage)
+
+        FIELD_RE = r'video\.%s\s*=\s*(["\'])(?P<url>(?:(?!\1).)+)\1'
+
+        media_url = self._html_search_regex(
+            FIELD_RE % 'src', webpage, 'video URL', group='url')
+
+        video_id = self._search_regex(
+            r'<link\b[^>]+\bhref=["\']https?://www\.ruv\.is/node/(\d+)',
+            webpage, 'video id', default=display_id)
+
+        ext = determine_ext(media_url)
+
+        if ext == 'm3u8':
+            formats = self._extract_m3u8_formats(
+                media_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                m3u8_id='hls')
+        elif ext == 'mp3':
+            formats = [{
+                'format_id': 'mp3',
+                'url': media_url,
+                'vcodec': 'none',
+            }]
+        else:
+            formats = [{
+                'url': media_url,
+            }]
+
+        description = self._og_search_description(webpage, default=None)
+        thumbnail = self._og_search_thumbnail(
+            webpage, default=None) or self._search_regex(
+            FIELD_RE % 'poster', webpage, 'thumbnail', fatal=False)
+        timestamp = unified_timestamp(self._html_search_meta(
+            'article:published_time', webpage, 'timestamp', fatal=False))
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'formats': formats,
+        }

youtube_dl/extractor/vimeo.py

@@ -615,7 +615,10 @@ class VimeoIE(VimeoBaseInfoExtractor):
                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
                     source_name = source_file.get('public_name', 'Original')
                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
-                        ext = source_file.get('extension', determine_ext(download_url)).lower()
+                        ext = (try_get(
+                            source_file, lambda x: x['extension'],
+                            compat_str) or determine_ext(
+                            download_url, None) or 'mp4').lower()
                         formats.append({
                             'url': download_url,
                             'ext': ext,

youtube_dl/extractor/viu.py

@@ -4,7 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_kwargs,
+    compat_str,
+)
 from ..utils import (
     ExtractorError,
     int_or_none,
@@ -36,7 +39,8 @@ class ViuBaseIE(InfoExtractor):
         headers.update(kwargs.get('headers', {}))
         kwargs['headers'] = headers
         response = self._download_json(
-            'https://www.viu.com/api/' + path, *args, **kwargs)['response']
+            'https://www.viu.com/api/' + path, *args,
+            **compat_kwargs(kwargs))['response']
         if response.get('status') != 'success':
             raise ExtractorError('%s said: %s' % (
                 self.IE_NAME, response['message']), expected=True)

youtube_dl/extractor/watchindianporn.py

@@ -4,11 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..utils import parse_duration
-    unified_strdate,
-    parse_duration,
-    int_or_none,
-)
 
 
 class WatchIndianPornIE(InfoExtractor):
@@ -23,11 +19,8 @@ class WatchIndianPornIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Hot milf from kerala shows off her gorgeous large breasts on camera',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'uploader': 'LoveJay',
-            'upload_date': '20160428',
             'duration': 226,
             'view_count': int,
-            'comment_count': int,
             'categories': list,
             'age_limit': 18,
         }
@@ -40,51 +33,36 @@ class WatchIndianPornIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        video_url = self._html_search_regex(
+        info_dict = self._parse_html5_media_entries(url, webpage, video_id)[0]
-            r"url: escape\('([^']+)'\)", webpage, 'url')
 
-        title = self._html_search_regex(
+        title = self._html_search_regex((
-            r'<h2 class="he2"><span>(.*?)</span>',
+            r'<title>(.+?)\s*-\s*Indian\s+Porn</title>',
-            webpage, 'title')
+            r'<h4>(.+?)</h4>'
-        thumbnail = self._html_search_regex(
+        ), webpage, 'title')
-            r'<span id="container"><img\s+src="([^"]+)"',
-            webpage, 'thumbnail', fatal=False)
-
-        uploader = self._html_search_regex(
-            r'class="aupa">\s*(.*?)</a>',
-            webpage, 'uploader')
-        upload_date = unified_strdate(self._html_search_regex(
-            r'Added: <strong>(.+?)</strong>', webpage, 'upload date', fatal=False))
 
         duration = parse_duration(self._search_regex(
-            r'<td>Time:\s*</td>\s*<td align="right"><span>\s*(.+?)\s*</span>',
+            r'Time:\s*<strong>\s*(.+?)\s*</strong>',
             webpage, 'duration', fatal=False))
 
-        view_count = int_or_none(self._search_regex(
+        view_count = int(self._search_regex(
-            r'<td>Views:\s*</td>\s*<td align="right"><span>\s*(\d+)\s*</span>',
+            r'(?s)Time:\s*<strong>.*?</strong>.*?<strong>\s*(\d+)\s*</strong>',
             webpage, 'view count', fatal=False))
-        comment_count = int_or_none(self._search_regex(
-            r'<td>Comments:\s*</td>\s*<td align="right"><span>\s*(\d+)\s*</span>',
-            webpage, 'comment count', fatal=False))
 
         categories = re.findall(
-            r'<a href="[^"]+/search/video/desi"><span>([^<]+)</span></a>',
+            r'<a[^>]+class=[\'"]categories[\'"][^>]*>\s*([^<]+)\s*</a>',
             webpage)
 
-        return {
+        info_dict.update({
             'id': video_id,
             'display_id': display_id,
-            'url': video_url,
             'http_headers': {
                 'Referer': url,
             },
             'title': title,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
-            'upload_date': upload_date,
             'duration': duration,
             'view_count': view_count,
-            'comment_count': comment_count,
             'categories': categories,
             'age_limit': 18,
-        }
+        })
+
+        return info_dict

youtube_dl/extractor/youporn.py

@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     int_or_none,
     sanitized_Request,
@@ -26,7 +27,7 @@ class YouPornIE(InfoExtractor):
             'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?',
             'thumbnail': r're:^https?://.*\.jpg$',
             'uploader': 'Ask Dan And Jennifer',
-            'upload_date': '20101221',
+            'upload_date': '20101217',
             'average_rating': int,
             'view_count': int,
             'comment_count': int,
@@ -45,7 +46,7 @@ class YouPornIE(InfoExtractor):
             'description': 'http://sweetlivegirls.com Big Tits Awesome Brunette On amazing webcam show.mp4',
             'thumbnail': r're:^https?://.*\.jpg$',
             'uploader': 'Unknown',
-            'upload_date': '20111125',
+            'upload_date': '20110418',
             'average_rating': int,
             'view_count': int,
             'comment_count': int,
@@ -68,28 +69,46 @@ class YouPornIE(InfoExtractor):
         webpage = self._download_webpage(request, display_id)
 
         title = self._search_regex(
-            [r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>.+?)\1',
+            [r'(?:video_titles|videoTitle)\s*[:=]\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
-             r'<h1[^>]+class=["\']heading\d?["\'][^>]*>([^<])<'],
+             r'<h1[^>]+class=["\']heading\d?["\'][^>]*>(?P<title>[^<]+)<'],
-            webpage, 'title', group='title')
+            webpage, 'title', group='title',
+            default=None) or self._og_search_title(
+            webpage, default=None) or self._html_search_meta(
+            'title', webpage, fatal=True)
 
         links = []
 
+        # Main source
+        definitions = self._parse_json(
+            self._search_regex(
+                r'mediaDefinition\s*=\s*(\[.+?\]);', webpage,
+                'media definitions', default='[]'),
+            video_id, fatal=False)
+        if definitions:
+            for definition in definitions:
+                if not isinstance(definition, dict):
+                    continue
+                video_url = definition.get('videoUrl')
+                if isinstance(video_url, compat_str) and video_url:
+                    links.append(video_url)
+
+        # Fallback #1, this also contains extra low quality 180p format
+        for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage):
+            links.append(link)
+
+        # Fallback #2 (unavailable as at 22.06.2017)
         sources = self._search_regex(
             r'(?s)sources\s*:\s*({.+?})', webpage, 'sources', default=None)
         if sources:
             for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources):
                 links.append(link)
 
-        # Fallback #1
+        # Fallback #3 (unavailable as at 22.06.2017)
         for _, link in re.findall(
-                r'(?:videoUrl|videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage):
+                r'(?:videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage):
             links.append(link)
 
-        # Fallback #2, this also contains extra low quality 180p format
+        # Fallback #4, encrypted links (unavailable as at 22.06.2017)
-        for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage):
-            links.append(link)
-
-        # Fallback #3, encrypted links
         for _, encrypted_link in re.findall(
                 r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage):
             links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8'))
@@ -124,7 +143,8 @@ class YouPornIE(InfoExtractor):
             r'(?s)<div[^>]+class=["\']submitByLink["\'][^>]*>(.+?)</div>',
             webpage, 'uploader', fatal=False)
         upload_date = unified_strdate(self._html_search_regex(
-            r'(?s)<div[^>]+class=["\']videoInfo(?:Date|Time)["\'][^>]*>(.+?)</div>',
+            [r'Date\s+[Aa]dded:\s*<span>([^<]+)',
+             r'(?s)<div[^>]+class=["\']videoInfo(?:Date|Time)["\'][^>]*>(.+?)</div>'],
             webpage, 'upload date', fatal=False))
 
         age_limit = self._rta_search(webpage)

youtube_dl/extractor/youtube.py

@@ -1269,37 +1269,57 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     sub_lang_list[sub_lang] = sub_formats
                 return sub_lang_list
 
-            # Some videos don't provide ttsurl but rather caption_tracks and
+            def make_captions(sub_url, sub_langs):
-            # caption_translation_languages (e.g. 20LmZk1hakA)
+                parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
-            caption_tracks = args['caption_tracks']
+                caption_qs = compat_parse_qs(parsed_sub_url.query)
-            caption_translation_languages = args['caption_translation_languages']
+                captions = {}
-            caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
+                for sub_lang in sub_langs:
-            parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
-            caption_qs = compat_parse_qs(parsed_caption_url.query)
-
-            sub_lang_list = {}
-            for lang in caption_translation_languages.split(','):
-                lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
-                sub_lang = lang_qs.get('lc', [None])[0]
-                if not sub_lang:
-                    continue
                     sub_formats = []
                     for ext in self._SUBTITLE_FORMATS:
                         caption_qs.update({
                             'tlang': [sub_lang],
                             'fmt': [ext],
                         })
-                    sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
+                        sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
                             query=compat_urllib_parse_urlencode(caption_qs, True)))
                         sub_formats.append({
                             'url': sub_url,
                             'ext': ext,
                         })
-                sub_lang_list[sub_lang] = sub_formats
+                    captions[sub_lang] = sub_formats
-            return sub_lang_list
+                return captions
+
+            # New captions format as of 22.06.2017
+            player_response = args.get('player_response')
+            if player_response and isinstance(player_response, compat_str):
+                player_response = self._parse_json(
+                    player_response, video_id, fatal=False)
+                if player_response:
+                    renderer = player_response['captions']['playerCaptionsTracklistRenderer']
+                    base_url = renderer['captionTracks'][0]['baseUrl']
+                    sub_lang_list = []
+                    for lang in renderer['translationLanguages']:
+                        lang_code = lang.get('languageCode')
+                        if lang_code:
+                            sub_lang_list.append(lang_code)
+                    return make_captions(base_url, sub_lang_list)
+
+            # Some videos don't provide ttsurl but rather caption_tracks and
+            # caption_translation_languages (e.g. 20LmZk1hakA)
+            # Does not used anymore as of 22.06.2017
+            caption_tracks = args['caption_tracks']
+            caption_translation_languages = args['caption_translation_languages']
+            caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
+            sub_lang_list = []
+            for lang in caption_translation_languages.split(','):
+                lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
+                sub_lang = lang_qs.get('lc', [None])[0]
+                if sub_lang:
+                    sub_lang_list.append(sub_lang)
+            return make_captions(caption_url, sub_lang_list)
         # An extractor error can be raise by the download process if there are
         # no automatic captions but there are subtitles
-        except (KeyError, ExtractorError):
+        except (KeyError, IndexError, ExtractorError):
             self._downloader.report_warning(err_msg)
             return {}
 

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2017.06.18'
+__version__ = '2017.06.23'