release 2017.07.23

All these meta fields must be borrowed from final extractor that actually performs extraction.
This commit fixes extractor id in download archives for url_transparent downloads. Previously, 'transparent' extractor was erroneously
used for extractor archive id, e.g. 'eggheadlesson 4n8ugwwj5t' instead of 'wistia 4n8ugwwj5t'.

.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 *2017.07.15*. 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 Issues with outdated version will be rejected.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 **2017.07.15**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.07.23*. 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 Issues with outdated version will be rejected.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 **2017.07.23**
 
 ### 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 2017.07.15
+[debug] youtube-dl version 2017.07.23
 [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 +1,30 @@
+version 2017.07.23
+
+Core
+* [YoutubeDL] Improve default format specification (#13704)
+* [YoutubeDL] Do not override id, extractor and extractor_key for
+  url_transparent entities
+* [extractor/common] Fix playlist_from_matches
+
+Extractors
+* [itv] Fix production id extraction (#13671, #13703)
+* [vidio] Make duration non fatal and fix typo
+* [mtv] Skip missing video parts (#13690)
+* [sportbox:embed] Fix extraction
++ [npo] Add support for npo3.nl URLs (#13695)
+* [dramafever] Remove video id from title (#13699)
++ [egghead:lesson] Add support for lessons (#6635)
+* [funnyordie] Extract more metadata (#13677)
+* [youku:show] Fix playlist extraction (#13248)
++ [dispeak] Recognize sevt subdomain (#13276)
+* [adn] Improve error reporting (#13663)
+* [crunchyroll] Relax series and season regex (#13659)
++ [spiegel:article] Add support for nexx iframe embeds (#13029)
++ [nexx:embed] Add support for iframe embeds
+* [nexx] Improve JS embed extraction
++ [pearvideo] Add support for pearvideo.com (#13031)
+
+
 version 2017.07.15
 
 Core

docs/supportedsites.md

@@ -42,7 +42,7 @@
  - **Allocine**
  - **AlphaPorno**
  - **AMCNetworks**
- - **anderetijden**: npo.nl and ntr.nl
+ - **anderetijden**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **AnimeOnDemand**
  - **anitube.se**
  - **Anvato**
@@ -238,6 +238,7 @@
  - **EbaumsWorld**
  - **EchoMsk**
  - **egghead:course**: egghead.io course
+ - **egghead:lesson**: egghead.io lesson
  - **eHow**
  - **Einthusan**
  - **eitb.tv**
@@ -522,6 +523,7 @@
  - **NextMediaActionNews**: 蘋果日報 - 動新聞
  - **NextTV**: 壹電視
  - **Nexx**
+ - **NexxEmbed**
  - **nfb**: National Film Board of Canada
  - **nfl.com**
  - **NhkVod**
@@ -552,7 +554,7 @@
  - **NowTVList**
  - **nowvideo**: NowVideo
  - **Noz**
- - **npo**: npo.nl and ntr.nl
+ - **npo**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **npo.nl:live**
  - **npo.nl:radio**
  - **npo.nl:radio:fragment**
@@ -596,6 +598,7 @@
  - **Patreon**
  - **pbs**: Public Broadcasting Service (PBS) and member stations: PBS: Public Broadcasting Service, APT - Alabama Public Television (WBIQ), GPB/Georgia Public Broadcasting (WGTV), Mississippi Public Broadcasting (WMPN), Nashville Public Television (WNPT), WFSU-TV (WFSU), WSRE (WSRE), WTCI (WTCI), WPBA/Channel 30 (WPBA), Alaska Public Media (KAKM), Arizona PBS (KAET), KNME-TV/Channel 5 (KNME), Vegas PBS (KLVX), AETN/ARKANSAS ETV NETWORK (KETS), KET (WKLE), WKNO/Channel 10 (WKNO), LPB/LOUISIANA PUBLIC BROADCASTING (WLPB), OETA (KETA), Ozarks Public Television (KOZK), WSIU Public Broadcasting (WSIU), KEET TV (KEET), KIXE/Channel 9 (KIXE), KPBS San Diego (KPBS), KQED (KQED), KVIE Public Television (KVIE), PBS SoCal/KOCE (KOCE), ValleyPBS (KVPT), CONNECTICUT PUBLIC TELEVISION (WEDH), KNPB Channel 5 (KNPB), SOPTV (KSYS), Rocky Mountain PBS (KRMA), KENW-TV3 (KENW), KUED Channel 7 (KUED), Wyoming PBS (KCWC), Colorado Public Television / KBDI 12 (KBDI), KBYU-TV (KBYU), Thirteen/WNET New York (WNET), WGBH/Channel 2 (WGBH), WGBY (WGBY), NJTV Public Media NJ (WNJT), WLIW21 (WLIW), mpt/Maryland Public Television (WMPB), WETA Television and Radio (WETA), WHYY (WHYY), PBS 39 (WLVT), WVPT - Your Source for PBS and More! (WVPT), Howard University Television (WHUT), WEDU PBS (WEDU), WGCU Public Media (WGCU), WPBT2 (WPBT), WUCF TV (WUCF), WUFT/Channel 5 (WUFT), WXEL/Channel 42 (WXEL), WLRN/Channel 17 (WLRN), WUSF Public Broadcasting (WUSF), ETV (WRLK), UNC-TV (WUNC), PBS Hawaii - Oceanic Cable Channel 10 (KHET), Idaho Public Television (KAID), KSPS (KSPS), OPB (KOPB), KWSU/Channel 10 & KTNW/Channel 31 (KWSU), WILL-TV (WILL), Network Knowledge - WSEC/Springfield (WSEC), WTTW11 (WTTW), Iowa Public Television/IPTV (KDIN), Nine Network (KETC), PBS39 Fort Wayne (WFWA), WFYI Indianapolis (WFYI), Milwaukee Public Television (WMVS), WNIN (WNIN), WNIT Public Television (WNIT), WPT (WPNE), WVUT/Channel 22 (WVUT), WEIU/Channel 51 (WEIU), WQPT-TV (WQPT), WYCC PBS Chicago (WYCC), WIPB-TV (WIPB), WTIU (WTIU), CET  (WCET), ThinkTVNetwork (WPTD), WBGU-TV (WBGU), WGVU TV (WGVU), NET1 (KUON), Pioneer Public Television (KWCM), SDPB Television (KUSD), TPT (KTCA), KSMQ (KSMQ), KPTS/Channel 8 (KPTS), KTWU/Channel 11 (KTWU), East Tennessee PBS (WSJK), WCTE-TV (WCTE), WLJT, Channel 11 (WLJT), WOSU TV (WOSU), WOUB/WOUC (WOUB), WVPB (WVPB), WKYU-PBS (WKYU), KERA 13 (KERA), MPBN (WCBB), Mountain Lake PBS (WCFE), NHPTV (WENH), Vermont PBS (WETK), witf (WITF), WQED Multimedia (WQED), WMHT Educational Telecommunications (WMHT), Q-TV (WDCQ), WTVS Detroit Public TV (WTVS), CMU Public Television (WCMU), WKAR-TV (WKAR), WNMU-TV Public TV 13 (WNMU), WDSE - WRPT (WDSE), WGTE TV (WGTE), Lakeland Public Television (KAWE), KMOS-TV - Channels 6.1, 6.2 and 6.3 (KMOS), MontanaPBS (KUSM), KRWG/Channel 22 (KRWG), KACV (KACV), KCOS/Channel 13 (KCOS), WCNY/Channel 24 (WCNY), WNED (WNED), WPBS (WPBS), WSKG Public TV (WSKG), WXXI (WXXI), WPSU (WPSU), WVIA Public Media Studios (WVIA), WTVI (WTVI), Western Reserve PBS (WNEO), WVIZ/PBS ideastream (WVIZ), KCTS 9 (KCTS), Basin PBS (KPBT), KUHT / Channel 8 (KUHT), KLRN (KLRN), KLRU (KLRU), WTJX Channel 12 (WTJX), WCVE PBS (WCVE), KBTC Public Television (KBTC)
  - **pcmag**
+ - **PearVideo**
  - **People**
  - **periscope**: Periscope
  - **periscope:user**: Periscope user videos
@@ -772,7 +775,7 @@
  - **tagesschau:player**
  - **Tass**
  - **TastyTrade**
- - **TBS**
+ - **TBS** (Currently broken)
  - **TDSLifeway**
  - **teachertube**: teachertube.com videos
  - **teachertube:user:collection**: teachertube.com user and collection videos
@@ -950,7 +953,7 @@
  - **VoiceRepublic**
  - **VoxMedia**
  - **Vporn**
- - **vpro**: npo.nl and ntr.nl
+ - **vpro**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **Vrak**
  - **VRT**: deredactie.be, sporza.be, cobra.be and cobra.canvas.be
  - **vrv**
@@ -976,7 +979,7 @@
  - **wholecloud**: WholeCloud
  - **Wimp**
  - **Wistia**
- - **wnl**: npo.nl and ntr.nl
+ - **wnl**: npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl
  - **WorldStarHipHop**
  - **wrzuta.pl**
  - **wrzuta.pl:playlist**

test/test_YoutubeDL.py

@@ -41,6 +41,7 @@ def _make_result(formats, **kwargs):
         'id': 'testid',
         'title': 'testttitle',
         'extractor': 'testex',
+        'extractor_key': 'TestEx',
     }
     res.update(**kwargs)
     return res
@@ -448,6 +449,17 @@ class TestFormatSelection(unittest.TestCase):
             pass
         self.assertEqual(ydl.downloaded_info_dicts, [])
 
+    def test_default_format_spec(self):
+        ydl = YDL({'simulate': True})
+        self.assertEqual(ydl._default_format_spec({}), 'bestvideo+bestaudio/best')
+
+        ydl = YDL({'outtmpl': '-'})
+        self.assertEqual(ydl._default_format_spec({}), 'best')
+
+        ydl = YDL({})
+        self.assertEqual(ydl._default_format_spec({}, download=False), 'bestvideo+bestaudio/best')
+        self.assertEqual(ydl._default_format_spec({'is_live': True}), 'best')
+
 
 class TestYoutubeDL(unittest.TestCase):
     def test_subtitles(self):
@@ -761,7 +773,8 @@ class TestYoutubeDL(unittest.TestCase):
                     '_type': 'url_transparent',
                     'url': 'foo2:',
                     'ie_key': 'Foo2',
-                    'title': 'foo1 title'
+                    'title': 'foo1 title',
+                    'id': 'foo1_id',
                 }
 
         class Foo2IE(InfoExtractor):
@@ -787,6 +800,9 @@ class TestYoutubeDL(unittest.TestCase):
         downloaded = ydl.downloaded_info_dicts[0]
         self.assertEqual(downloaded['url'], TEST_URL)
         self.assertEqual(downloaded['title'], 'foo1 title')
+        self.assertEqual(downloaded['id'], 'testid')
+        self.assertEqual(downloaded['extractor'], 'testex')
+        self.assertEqual(downloaded['extractor_key'], 'TestEx')
 
 
 if __name__ == '__main__':

youtube_dl/YoutubeDL.py

@@ -860,7 +860,7 @@ class YoutubeDL(object):
 
             force_properties = dict(
                 (k, v) for k, v in ie_result.items() if v is not None)
-            for f in ('_type', 'url', 'ie_key'):
+            for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
                 if f in force_properties:
                     del force_properties[f]
             new_result = info.copy()
@@ -1064,6 +1064,25 @@ class YoutubeDL(object):
             return op(actual_value, comparison_value)
         return _filter
 
+    def _default_format_spec(self, info_dict, download=True):
+        req_format_list = []
+
+        def can_have_partial_formats():
+            if self.params.get('simulate', False):
+                return True
+            if not download:
+                return True
+            if self.params.get('outtmpl', DEFAULT_OUTTMPL) == '-':
+                return False
+            if info_dict.get('is_live'):
+                return False
+            merger = FFmpegMergerPP(self)
+            return merger.available and merger.can_merge()
+        if can_have_partial_formats():
+            req_format_list.append('bestvideo+bestaudio')
+        req_format_list.append('best')
+        return '/'.join(req_format_list)
+
     def build_format_selector(self, format_spec):
         def syntax_error(note, start):
             message = (
@@ -1534,14 +1553,10 @@ class YoutubeDL(object):
 
         req_format = self.params.get('format')
         if req_format is None:
-            req_format_list = []
-            if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
-                    not info_dict.get('is_live')):
-                merger = FFmpegMergerPP(self)
-                if merger.available and merger.can_merge():
-                    req_format_list.append('bestvideo+bestaudio')
-            req_format_list.append('best')
-            req_format = '/'.join(req_format_list)
+            req_format = self._default_format_spec(info_dict, download=download)
+            if self.params.get('verbose'):
+                self.to_stdout('[debug] Default format spec: %s' % req_format)
+
         format_selector = self.build_format_selector(req_format)
 
         # While in format selection we may need to have an access to the original

youtube_dl/extractor/adn.py

@@ -107,11 +107,13 @@ class ADNIE(InfoExtractor):
         metas = options.get('metas') or {}
         title = metas.get('title') or video_info['title']
         links = player_config.get('links') or {}
+        error = None
         if not links:
             links_url = player_config['linksurl']
             links_data = self._download_json(urljoin(
                 self._BASE_URL, links_url), video_id)
             links = links_data.get('links') or {}
+            error = links_data.get('error')
 
         formats = []
         for format_id, qualities in links.items():
@@ -130,7 +132,8 @@ class ADNIE(InfoExtractor):
                     for f in m3u8_formats:
                         f['language'] = 'fr'
                 formats.extend(m3u8_formats)
-        error = options.get('error')
+        if not error:
+            error = options.get('error')
         if not formats and error:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
         self._sort_formats(formats)

youtube_dl/extractor/common.py

@@ -730,12 +730,12 @@ class InfoExtractor(object):
             video_info['title'] = video_title
         return video_info
 
-    def playlist_from_matches(self, matches, video_id, video_title, getter=None, ie=None):
-        urlrs = orderedSet(
+    def playlist_from_matches(self, matches, playlist_id=None, playlist_title=None, getter=None, ie=None):
+        urls = orderedSet(
             self.url_result(self._proto_relative_url(getter(m) if getter else m), ie)
             for m in matches)
         return self.playlist_result(
-            urlrs, playlist_id=video_id, playlist_title=video_title)
+            urls, playlist_id=playlist_id, playlist_title=playlist_title)
 
     @staticmethod
     def playlist_result(entries, playlist_id=None, playlist_title=None, playlist_description=None):

youtube_dl/extractor/crunchyroll.py

@@ -510,7 +510,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
         # webpage provide more accurate data than series_title from XML
         series = self._html_search_regex(
-            r'id=["\']showmedia_about_episode_num[^>]+>\s*<a[^>]+>([^<]+)',
+            r'(?s)<h\d[^>]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)</h\d',
             webpage, 'series', fatal=False)
         season = xpath_text(metadata, 'series_title')
 
@@ -518,7 +518,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
 
         season_number = int_or_none(self._search_regex(
-            r'(?s)<h4[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h4>\s*<h4>\s*Season (\d+)',
+            r'(?s)<h\d[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h\d>\s*<h4>\s*Season (\d+)',
             webpage, 'season number', default=None))
 
         return {

youtube_dl/extractor/dispeak.py

@@ -13,7 +13,7 @@ from ..utils import (
 
 
 class DigitallySpeakingIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:evt\.dispeak|events\.digitallyspeaking)\.com/(?:[^/]+/)+xml/(?P<id>[^.]+)\.xml'
+    _VALID_URL = r'https?://(?:s?evt\.dispeak|events\.digitallyspeaking)\.com/(?:[^/]+/)+xml/(?P<id>[^.]+)\.xml'
 
     _TESTS = [{
         # From http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface
@@ -28,6 +28,10 @@ class DigitallySpeakingIE(InfoExtractor):
         # From http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC
         'url': 'http://events.digitallyspeaking.com/gdc/sf11/xml/12396_1299111843500GMPX.xml',
         'only_matching': True,
+    }, {
+        # From http://www.gdcvault.com/play/1013700/Advanced-Material
+        'url': 'http://sevt.dispeak.com/ubm/gdc/eur10/xml/11256_1282118587281VNIT.xml',
+        'only_matching': True,
     }]
 
     def _parse_mp4(self, metadata):

youtube_dl/extractor/dramafever.py

@@ -12,6 +12,7 @@ from ..utils import (
     ExtractorError,
     clean_html,
     int_or_none,
+    remove_end,
     sanitized_Request,
     urlencode_postdata
 )
@@ -72,15 +73,15 @@ class DramaFeverIE(DramaFeverBaseIE):
         'url': 'http://www.dramafever.com/drama/4512/1/Cooking_with_Shin/',
         'info_dict': {
             'id': '4512.1',
-            'ext': 'mp4',
-            'title': 'Cooking with Shin 4512.1',
+            'ext': 'flv',
+            'title': 'Cooking with Shin',
             'description': 'md5:a8eec7942e1664a6896fcd5e1287bfd0',
             'episode': 'Episode 1',
             'episode_number': 1,
             'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1404336058,
             'upload_date': '20140702',
-            'duration': 343,
+            'duration': 344,
         },
         'params': {
             # m3u8 download
@@ -90,15 +91,15 @@ class DramaFeverIE(DramaFeverBaseIE):
         'url': 'http://www.dramafever.com/drama/4826/4/Mnet_Asian_Music_Awards_2015/?ap=1',
         'info_dict': {
             'id': '4826.4',
-            'ext': 'mp4',
-            'title': 'Mnet Asian Music Awards 2015 4826.4',
+            'ext': 'flv',
+            'title': 'Mnet Asian Music Awards 2015',
             'description': 'md5:3ff2ee8fedaef86e076791c909cf2e91',
             'episode': 'Mnet Asian Music Awards 2015 - Part 3',
             'episode_number': 4,
             'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1450213200,
             'upload_date': '20151215',
-            'duration': 5602,
+            'duration': 5359,
         },
         'params': {
             # m3u8 download
@@ -122,6 +123,10 @@ class DramaFeverIE(DramaFeverBaseIE):
                     countries=self._GEO_COUNTRIES)
             raise
 
+        # title is postfixed with video id for some reason, removing
+        if info.get('title'):
+            info['title'] = remove_end(info['title'], video_id).strip()
+
         series_id, episode_number = video_id.split('.')
         episode_info = self._download_json(
             # We only need a single episode info, so restricting page size to one episode

youtube_dl/extractor/egghead.py

@@ -2,6 +2,11 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    try_get,
+    unified_timestamp,
+)
 
 
 class EggheadCourseIE(InfoExtractor):
@@ -33,3 +38,47 @@ class EggheadCourseIE(InfoExtractor):
         return self.playlist_result(
             entries, playlist_id, course.get('title'),
             course.get('description'))
+
+
+class EggheadLessonIE(InfoExtractor):
+    IE_DESC = 'egghead.io lesson'
+    IE_NAME = 'egghead:lesson'
+    _VALID_URL = r'https://egghead\.io/lessons/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box',
+        'info_dict': {
+            'id': 'fv5yotjxcg',
+            'ext': 'mp4',
+            'title': 'Create linear data flow with container style types (Box)',
+            'description': 'md5:9aa2cdb6f9878ed4c39ec09e85a8150e',
+            'thumbnail': r're:^https?:.*\.jpg$',
+            'timestamp': 1481296768,
+            'upload_date': '20161209',
+            'duration': 304,
+            'view_count': 0,
+            'tags': ['javascript', 'free'],
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        lesson_id = self._match_id(url)
+
+        lesson = self._download_json(
+            'https://egghead.io/api/v1/lessons/%s' % lesson_id, lesson_id)
+
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'Wistia',
+            'url': 'wistia:%s' % lesson['wistia_id'],
+            'id': lesson['wistia_id'],
+            'title': lesson.get('title'),
+            'description': lesson.get('summary'),
+            'thumbnail': lesson.get('thumb_nail'),
+            'timestamp': unified_timestamp(lesson.get('published_at')),
+            'duration': int_or_none(lesson.get('duration')),
+            'view_count': int_or_none(lesson.get('plays_count')),
+            'tags': try_get(lesson, lambda x: x['tag_list'], list),
+        }

youtube_dl/extractor/extractors.py

@@ -298,7 +298,10 @@ from .dw import (
 from .eagleplatform import EaglePlatformIE
 from .ebaumsworld import EbaumsWorldIE
 from .echomsk import EchoMskIE
-from .egghead import EggheadCourseIE
+from .egghead import (
+    EggheadCourseIE,
+    EggheadLessonIE,
+)
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .einthusan import EinthusanIE
@@ -653,7 +656,10 @@ from .nextmedia import (
     AppleDailyIE,
     NextTVIE,
 )
-from .nexx import NexxIE
+from .nexx import (
+    NexxIE,
+    NexxEmbedIE,
+)
 from .nfb import NFBIE
 from .nfl import NFLIE
 from .nhk import NhkVodIE
@@ -762,6 +768,7 @@ from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE
+from .pearvideo import PearVideoIE
 from .people import PeopleIE
 from .periscope import (
     PeriscopeIE,

youtube_dl/extractor/funnyordie.py

@@ -1,10 +1,14 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    float_or_none,
+    int_or_none,
+    unified_timestamp,
+)
 
 
 class FunnyOrDieIE(InfoExtractor):
@@ -18,6 +22,10 @@ class FunnyOrDieIE(InfoExtractor):
             'title': 'Heart-Shaped Box: Literal Video Version',
             'description': 'md5:ea09a01bc9a1c46d9ab696c01747c338',
             'thumbnail': r're:^http:.*\.jpg$',
+            'uploader': 'DASjr',
+            'timestamp': 1317904928,
+            'upload_date': '20111006',
+            'duration': 318.3,
         },
     }, {
         'url': 'http://www.funnyordie.com/embed/e402820827',
@@ -27,6 +35,8 @@ class FunnyOrDieIE(InfoExtractor):
             'title': 'Please Use This Song (Jon Lajoie)',
             'description': 'Please use this to sell something.  www.jonlajoie.com',
             'thumbnail': r're:^http:.*\.jpg$',
+            'timestamp': 1398988800,
+            'upload_date': '20140502',
         },
         'params': {
             'skip_download': True,
@@ -100,15 +110,53 @@ class FunnyOrDieIE(InfoExtractor):
                 'url': 'http://www.funnyordie.com%s' % src,
             }]
 
-        post_json = self._search_regex(
-            r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details')
-        post = json.loads(post_json)
+        timestamp = unified_timestamp(self._html_search_meta(
+            'uploadDate', webpage, 'timestamp', default=None))
+
+        uploader = self._html_search_regex(
+            r'<h\d[^>]+\bclass=["\']channel-preview-name[^>]+>(.+?)</h',
+            webpage, 'uploader', default=None)
+
+        title, description, thumbnail, duration = [None] * 4
+
+        medium = self._parse_json(
+            self._search_regex(
+                r'jsonMedium\s*=\s*({.+?});', webpage, 'JSON medium',
+                default='{}'),
+            video_id, fatal=False)
+        if medium:
+            title = medium.get('title')
+            duration = float_or_none(medium.get('duration'))
+            if not timestamp:
+                timestamp = unified_timestamp(medium.get('publishDate'))
+
+        post = self._parse_json(
+            self._search_regex(
+                r'fb_post\s*=\s*(\{.*?\});', webpage, 'post details',
+                default='{}'),
+            video_id, fatal=False)
+        if post:
+            if not title:
+                title = post.get('name')
+            description = post.get('description')
+            thumbnail = post.get('picture')
+
+        if not title:
+            title = self._og_search_title(webpage)
+        if not description:
+            description = self._og_search_description(webpage)
+        if not duration:
+            duration = int_or_none(self._html_search_meta(
+                ('video:duration', 'duration'), webpage, 'duration', default=False))
 
         return {
             'id': video_id,
-            'title': post['name'],
-            'description': post.get('description'),
-            'thumbnail': post.get('picture'),
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'timestamp': timestamp,
+            'duration': duration,
             'formats': formats,
             'subtitles': subtitles,
         }

youtube_dl/extractor/generic.py

@@ -36,7 +36,10 @@ from .brightcove import (
     BrightcoveLegacyIE,
     BrightcoveNewIE,
 )
-from .nexx import NexxIE
+from .nexx import (
+    NexxIE,
+    NexxEmbedIE,
+)
 from .nbc import NBCSportsVPlayerIE
 from .ooyala import OoyalaIE
 from .rutv import RUTVIE
@@ -2155,6 +2158,11 @@ class GenericIE(InfoExtractor):
         if nexx_urls:
             return self.playlist_from_matches(nexx_urls, video_id, video_title, ie=NexxIE.ie_key())
 
+        # Look for Nexx iFrame embeds
+        nexx_embed_urls = NexxEmbedIE._extract_urls(webpage)
+        if nexx_embed_urls:
+            return self.playlist_from_matches(nexx_embed_urls, video_id, video_title, ie=NexxEmbedIE.ie_key())
+
         # Look for ThePlatform embeds
         tp_urls = ThePlatformIE._extract_urls(webpage)
         if tp_urls:

youtube_dl/extractor/itv.py

@@ -59,12 +59,18 @@ class ITVIE(InfoExtractor):
         def _add_sub_element(element, name):
             return etree.SubElement(element, _add_ns(name))
 
+        production_id = (
+            params.get('data-video-autoplay-id') or
+            '%s#001' % (
+                params.get('data-video-episode-id') or
+                video_id.replace('a', '/')))
+
         req_env = etree.Element(_add_ns('soapenv:Envelope'))
         _add_sub_element(req_env, 'soapenv:Header')
         body = _add_sub_element(req_env, 'soapenv:Body')
         get_playlist = _add_sub_element(body, ('tem:GetPlaylist'))
         request = _add_sub_element(get_playlist, 'tem:request')
-        _add_sub_element(request, 'itv:ProductionId').text = params['data-video-id']
+        _add_sub_element(request, 'itv:ProductionId').text = production_id
         _add_sub_element(request, 'itv:RequestGuid').text = compat_str(uuid.uuid4()).upper()
         vodcrid = _add_sub_element(request, 'itv:Vodcrid')
         _add_sub_element(vodcrid, 'com:Id')

youtube_dl/extractor/mtv.py

@@ -83,7 +83,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
                 hls_url = rendition.find('./src').text
                 formats.extend(self._extract_m3u8_formats(
                     hls_url, video_id, ext='mp4', entry_protocol='m3u8_native',
-                    m3u8_id='hls'))
+                    m3u8_id='hls', fatal=False))
             else:
                 # fms
                 try:
@@ -106,7 +106,8 @@ class MTVServicesInfoExtractor(InfoExtractor):
                     }])
                 except (KeyError, TypeError):
                     raise ExtractorError('Invalid rendition field.')
-        self._sort_formats(formats)
+        if formats:
+            self._sort_formats(formats)
         return formats
 
     def _extract_subtitles(self, mdoc, mtvn_id):
@@ -133,8 +134,11 @@ class MTVServicesInfoExtractor(InfoExtractor):
             mediagen_url += 'acceptMethods='
             mediagen_url += 'hls' if use_hls else 'fms'
 
-        mediagen_doc = self._download_xml(mediagen_url, video_id,
-                                          'Downloading video urls')
+        mediagen_doc = self._download_xml(
+            mediagen_url, video_id, 'Downloading video urls', fatal=False)
+
+        if mediagen_doc is False:
+            return None
 
         item = mediagen_doc.find('./video/item')
         if item is not None and item.get('type') == 'text':
@@ -174,6 +178,13 @@ class MTVServicesInfoExtractor(InfoExtractor):
 
         formats = self._extract_video_formats(mediagen_doc, mtvn_id, video_id)
 
+        # Some parts of complete video may be missing (e.g. missing Act 3 in
+        # http://www.southpark.de/alle-episoden/s14e01-sexual-healing)
+        if not formats:
+            return None
+
+        self._sort_formats(formats)
+
         return {
             'title': title,
             'formats': formats,
@@ -205,9 +216,14 @@ class MTVServicesInfoExtractor(InfoExtractor):
         title = xpath_text(idoc, './channel/title')
         description = xpath_text(idoc, './channel/description')
 
+        entries = []
+        for item in idoc.findall('.//item'):
+            info = self._get_video_info(item, use_hls)
+            if info:
+                entries.append(info)
+
         return self.playlist_result(
-            [self._get_video_info(item, use_hls) for item in idoc.findall('.//item')],
-            playlist_title=title, playlist_description=description)
+            entries, playlist_title=title, playlist_description=description)
 
     def _extract_triforce_mgid(self, webpage, data_zone=None, video_id=None):
         triforce_feed = self._parse_json(self._search_regex(

youtube_dl/extractor/nexx.py

@@ -72,18 +72,26 @@ class NexxIE(InfoExtractor):
         entries = []
 
         # JavaScript Integration
-        for domain_id, video_id in re.findall(
-                r'''(?isx)
-                    <script\b[^>]+\bsrc=["\']https?://require\.nexx(?:\.cloud|cdn\.com)/(\d+).+?
-                    onPLAYReady.+?
-                    _play\.init\s*\(.+?\s*,\s*(\d+)\s*,\s*.+?\)
-                ''', webpage):
-            entries.append('https://api.nexx.cloud/v3/%s/videos/byid/%s' % (domain_id, video_id))
+        mobj = re.search(
+            r'<script\b[^>]+\bsrc=["\']https?://require\.nexx(?:\.cloud|cdn\.com)/(?P<id>\d+)',
+            webpage)
+        if mobj:
+            domain_id = mobj.group('id')
+            for video_id in re.findall(
+                    r'(?is)onPLAYReady.+?_play\.init\s*\(.+?\s*,\s*["\']?(\d+)',
+                    webpage):
+                entries.append(
+                    'https://api.nexx.cloud/v3/%s/videos/byid/%s'
+                    % (domain_id, video_id))
 
         # TODO: support more embed formats
 
         return entries
 
+    @staticmethod
+    def _extract_url(webpage):
+        return NexxIE._extract_urls(webpage)[0]
+
     def _handle_error(self, response):
         status = int_or_none(try_get(
             response, lambda x: x['metadata']['status']) or 200)
@@ -219,3 +227,45 @@ class NexxIE(InfoExtractor):
                 video, lambda x: x['episodedata']['season'])),
             'formats': formats,
         }
+
+
+class NexxEmbedIE(InfoExtractor):
+    _VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'http://embed.nexx.cloud/748/KC1614647Z27Y7T?autoplay=1',
+        'md5': '16746bfc28c42049492385c989b26c4a',
+        'info_dict': {
+            'id': '161464',
+            'ext': 'mp4',
+            'title': 'Nervenkitzel Achterbahn',
+            'alt_title': 'Karussellbauer in Deutschland',
+            'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+            'release_year': 2005,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2761,
+            'timestamp': 1394021479,
+            'upload_date': '20140305',
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+    }
+
+    @staticmethod
+    def _extract_urls(webpage):
+        # Reference:
+        # 1. https://nx-s.akamaized.net/files/201510/44.pdf
+
+        # iFrame Embed Integration
+        return [mobj.group('url') for mobj in re.finditer(
+            r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?:(?!\1).)+)\1',
+            webpage)]
+
+    def _real_extract(self, url):
+        embed_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, embed_id)
+
+        return self.url_result(NexxIE._extract_url(webpage), ie=NexxIE.ie_key())

youtube_dl/extractor/npo.py

@@ -28,7 +28,7 @@ class NPOBaseIE(InfoExtractor):
 
 class NPOIE(NPOBaseIE):
     IE_NAME = 'npo'
-    IE_DESC = 'npo.nl and ntr.nl'
+    IE_DESC = 'npo.nl, ntr.nl, omroepwnl.nl, zapp.nl and npo3.nl'
     _VALID_URL = r'''(?x)
                     (?:
                         npo:|
@@ -38,7 +38,7 @@ class NPOIE(NPOBaseIE):
                                 npo\.nl/(?!(?:live|radio)/)(?:[^/]+/){2}|
                                 ntr\.nl/(?:[^/]+/){2,}|
                                 omroepwnl\.nl/video/fragment/[^/]+__|
-                                zapp\.nl/[^/]+/[^/]+/
+                                (?:zapp|npo3)\.nl/(?:[^/]+/){2}
                             )
                         )
                         (?P<id>[^/?#]+)
@@ -146,6 +146,9 @@ class NPOIE(NPOBaseIE):
     }, {
         'url': 'http://www.zapp.nl/beste-vrienden-quiz/extra-video-s/WO_NTR_1067990',
         'only_matching': True,
+    }, {
+        'url': 'https://www.npo3.nl/3onderzoekt/16-09-2015/VPWON_1239870',
+        'only_matching': True,
     }, {
         # live stream
         'url': 'npo:LI_NL1_4188102',

youtube_dl/extractor/pearvideo.py

@@ -0,0 +1,63 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    qualities,
+    unified_timestamp,
+)
+
+
+class PearVideoIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.pearvideo.com/video_1076290',
+        'info_dict': {
+            'id': '1076290',
+            'ext': 'mp4',
+            'title': '小浣熊在主人家玻璃上滚石头：没砸',
+            'description': 'md5:01d576b747de71be0ee85eb7cac25f9d',
+            'timestamp': 1494275280,
+            'upload_date': '20170508',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        quality = qualities(
+            ('ldflv', 'ld', 'sdflv', 'sd', 'hdflv', 'hd', 'src'))
+
+        formats = [{
+            'url': mobj.group('url'),
+            'format_id': mobj.group('id'),
+            'quality': quality(mobj.group('id')),
+        } for mobj in re.finditer(
+            r'(?P<id>[a-zA-Z]+)Url\s*=\s*(["\'])(?P<url>(?:https?:)?//.+?)\2',
+            webpage)]
+        self._sort_formats(formats)
+
+        title = self._search_regex(
+            (r'<h1[^>]+\bclass=(["\'])video-tt\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-title=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'title', group='value')
+        description = self._search_regex(
+            (r'<div[^>]+\bclass=(["\'])summary\1[^>]*>(?P<value>[^<]+)',
+             r'<[^>]+\bdata-summary=(["\'])(?P<value>(?:(?!\1).)+)\1'),
+            webpage, 'description', default=None,
+            group='value') or self._html_search_meta('Description', webpage)
+        timestamp = unified_timestamp(self._search_regex(
+            r'<div[^>]+\bclass=["\']date["\'][^>]*>([^<]+)',
+            webpage, 'timestamp', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'formats': formats,
+        }

youtube_dl/extractor/spiegel.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from .nexx import NexxEmbedIE
 from .spiegeltv import SpiegeltvIE
 from ..compat import compat_urlparse
 from ..utils import (
@@ -121,6 +122,26 @@ class SpiegelArticleIE(InfoExtractor):
 
         },
         'playlist_count': 6,
+    }, {
+        # Nexx iFrame embed
+        'url': 'http://www.spiegel.de/sptv/spiegeltv/spiegel-tv-ueber-schnellste-katapult-achterbahn-der-welt-taron-a-1137884.html',
+        'info_dict': {
+            'id': '161464',
+            'ext': 'mp4',
+            'title': 'Nervenkitzel Achterbahn',
+            'alt_title': 'Karussellbauer in Deutschland',
+            'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
+            'release_year': 2005,
+            'creator': 'SPIEGEL TV',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2761,
+            'timestamp': 1394021479,
+            'upload_date': '20140305',
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
@@ -143,6 +164,9 @@ class SpiegelArticleIE(InfoExtractor):
         entries = [
             self.url_result(compat_urlparse.urljoin(
                 self.http_scheme() + '//spiegel.de/', embed_path))
-            for embed_path in embeds
-        ]
-        return self.playlist_result(entries)
+            for embed_path in embeds]
+        if embeds:
+            return self.playlist_result(entries)
+
+        return self.playlist_from_matches(
+            NexxEmbedIE._extract_urls(webpage), ie=NexxEmbedIE.ie_key())

youtube_dl/extractor/sportbox.py

@@ -4,7 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import js_to_json
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    js_to_json,
+)
 
 
 class SportBoxEmbedIE(InfoExtractor):
@@ -14,8 +18,10 @@ class SportBoxEmbedIE(InfoExtractor):
         'info_dict': {
             'id': '211355',
             'ext': 'mp4',
-            'title': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
+            'title': '211355',
             'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 292,
+            'view_count': int,
         },
         'params': {
             # m3u8 download
@@ -24,6 +30,9 @@ class SportBoxEmbedIE(InfoExtractor):
     }, {
         'url': 'http://news.sportbox.ru/vdl/player?nid=370908&only_player=1&autostart=false&playeri=2&height=340&width=580',
         'only_matching': True,
+    }, {
+        'url': 'https://news.sportbox.ru/vdl/player/media/193095',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -37,36 +46,34 @@ class SportBoxEmbedIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        wjplayer_data = self._parse_json(
+            self._search_regex(
+                r'(?s)wjplayer\(({.+?})\);', webpage, 'wjplayer settings'),
+            video_id, transform_source=js_to_json)
+
         formats = []
-
-        def cleanup_js(code):
-            # desktop_advert_config contains complex Javascripts and we don't need it
-            return js_to_json(re.sub(r'desktop_advert_config.*', '', code))
-
-        jwplayer_data = self._parse_json(self._search_regex(
-            r'(?s)player\.setup\(({.+?})\);', webpage, 'jwplayer settings'), video_id,
-            transform_source=cleanup_js)
-
-        hls_url = jwplayer_data.get('hls_url')
-        if hls_url:
-            formats.extend(self._extract_m3u8_formats(
-                hls_url, video_id, ext='mp4', m3u8_id='hls'))
-
-        rtsp_url = jwplayer_data.get('rtsp_url')
-        if rtsp_url:
-            formats.append({
-                'url': rtsp_url,
-                'format_id': 'rtsp',
-            })
-
+        for source in wjplayer_data['sources']:
+            src = source.get('src')
+            if not src:
+                continue
+            if determine_ext(src) == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    src, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'url': src,
+                })
         self._sort_formats(formats)
 
-        title = jwplayer_data['node_title']
-        thumbnail = jwplayer_data.get('image_url')
+        view_count = int_or_none(self._search_regex(
+            r'Просмотров\s*:\s*(\d+)', webpage, 'view count', default=None))
 
         return {
             'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
+            'title': video_id,
+            'thumbnail': wjplayer_data.get('poster'),
+            'duration': int_or_none(wjplayer_data.get('duration')),
+            'view_count': view_count,
             'formats': formats,
         }

youtube_dl/extractor/tbs.py

@@ -8,6 +8,9 @@ from ..utils import extract_attributes
 
 
 class TBSIE(TurnerBaseIE):
+    # https://github.com/rg3/youtube-dl/issues/13658
+    _WORKING = False
+
     _VALID_URL = r'https?://(?:www\.)?(?P<site>tbs|tntdrama)\.com/videos/(?:[^/]+/)+(?P<id>[^/?#]+)\.html'
     _TESTS = [{
         'url': 'http://www.tbs.com/videos/people-of-earth/season-1/extras/2007318/theatrical-trailer.html',
@@ -17,7 +20,8 @@ class TBSIE(TurnerBaseIE):
             'ext': 'mp4',
             'title': 'Theatrical Trailer',
             'description': 'Catch the latest comedy from TBS, People of Earth, premiering Halloween night--Monday, October 31, at 9/8c.',
-        }
+        },
+        'skip': 'TBS videos are deleted after a while',
     }, {
         'url': 'http://www.tntdrama.com/videos/good-behavior/season-1/extras/1538823/you-better-run.html',
         'md5': 'ce53c6ead5e9f3280b4ad2031a6fab56',
@@ -26,7 +30,8 @@ class TBSIE(TurnerBaseIE):
             'ext': 'mp4',
             'title': 'You Better Run',
             'description': 'Letty Raines must figure out what she\'s running toward while running away from her past. Good Behavior premieres November 15 at 9/8c.',
-        }
+        },
+        'skip': 'TBS videos are deleted after a while',
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/vh1.py

@@ -121,7 +121,11 @@ class VH1IE(MTVIE):
         idoc = self._download_xml(
             doc_url, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
-        return self.playlist_result(
-            [self._get_video_info(item) for item in idoc.findall('.//item')],
-            playlist_id=video_id,
-        )
+
+        entries = []
+        for item in idoc.findall('.//item'):
+            info = self._get_video_info(item)
+            if info:
+                entries.append(info)
+
+        return self.playlist_result(entries, playlist_id=video_id)

youtube_dl/extractor/vidio.py

@@ -56,7 +56,8 @@ class VidioIE(InfoExtractor):
         self._sort_formats(formats)
 
         duration = int_or_none(duration or self._search_regex(
-            r'data-video-duration=(["\'])(?P<duartion>\d+)\1', webpage, 'duration'))
+            r'data-video-duration=(["\'])(?P<duration>\d+)\1', webpage,
+            'duration', fatal=False, group='duration'))
         thumbnail = thumbnail or self._og_search_thumbnail(webpage)
 
         like_count = int_or_none(self._search_regex(

youtube_dl/extractor/youku.py

@@ -1,7 +1,6 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import itertools
 import random
 import re
 import string
@@ -14,7 +13,6 @@ from ..utils import (
     js_to_json,
     str_or_none,
     strip_jsonp,
-    urljoin,
 )
 
 
@@ -222,17 +220,42 @@ class YoukuShowIE(InfoExtractor):
     _VALID_URL = r'https?://list\.youku\.com/show/id_(?P<id>[0-9a-z]+)\.html'
     IE_NAME = 'youku:show'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://list.youku.com/show/id_zc7c670be07ff11e48b3f.html',
         'info_dict': {
             'id': 'zc7c670be07ff11e48b3f',
-            'title': '花千骨 未删减版',
+            'title': '花千骨 DVD版',
             'description': 'md5:a1ae6f5618571bbeb5c9821f9c81b558',
         },
         'playlist_count': 50,
-    }
+    }, {
+        # Episode number not starting from 1
+        'url': 'http://list.youku.com/show/id_zefbfbd70efbfbd780bef.html',
+        'info_dict': {
+            'id': 'zefbfbd70efbfbd780bef',
+            'title': '超级飞侠3',
+            'description': 'md5:275715156abebe5ccc2a1992e9d56b98',
+        },
+        'playlist_count': 24,
+    }, {
+        # Ongoing playlist. The initial page is the last one
+        'url': 'http://list.youku.com/show/id_za7c275ecd7b411e1a19e.html',
+        'only_matchine': True,
+    }]
 
-    _PAGE_SIZE = 40
+    def _extract_entries(self, playlist_data_url, show_id, note, query):
+        query['callback'] = 'cb'
+        playlist_data = self._download_json(
+            playlist_data_url, show_id, query=query, note=note,
+            transform_source=lambda s: js_to_json(strip_jsonp(s)))['html']
+        drama_list = (get_element_by_class('p-drama-grid', playlist_data) or
+                      get_element_by_class('p-drama-half-row', playlist_data))
+        if drama_list is None:
+            raise ExtractorError('No episodes found')
+        video_urls = re.findall(r'<a[^>]+href="([^"]+)"', drama_list)
+        return playlist_data, [
+            self.url_result(self._proto_relative_url(video_url, 'http:'), YoukuIE.ie_key())
+            for video_url in video_urls]
 
     def _real_extract(self, url):
         show_id = self._match_id(url)
@@ -242,30 +265,29 @@ class YoukuShowIE(InfoExtractor):
         page_config = self._parse_json(self._search_regex(
             r'var\s+PageConfig\s*=\s*({.+});', webpage, 'page config'),
             show_id, transform_source=js_to_json)
-        for idx in itertools.count(0):
-            if idx == 0:
-                playlist_data_url = 'http://list.youku.com/show/module'
-                query = {'id': page_config['showid'], 'tab': 'point'}
-            else:
-                playlist_data_url = 'http://list.youku.com/show/point'
-                query = {
-                    'id': page_config['showid'],
-                    'stage': 'reload_%d' % (self._PAGE_SIZE * idx + 1),
-                }
-            query['callback'] = 'cb'
-            playlist_data = self._download_json(
-                playlist_data_url, show_id, query=query,
+        first_page, initial_entries = self._extract_entries(
+            'http://list.youku.com/show/module', show_id,
+            note='Downloading initial playlist data page',
+            query={
+                'id': page_config['showid'],
+                'tab': 'showInfo',
+            })
+        first_page_reload_id = self._html_search_regex(
+            r'<div[^>]+id="(reload_\d+)', first_page, 'first page reload id')
+        # The first reload_id has the same items as first_page
+        reload_ids = re.findall('<li[^>]+data-id="([^"]+)">', first_page)
+        for idx, reload_id in enumerate(reload_ids):
+            if reload_id == first_page_reload_id:
+                entries.extend(initial_entries)
+                continue
+            _, new_entries = self._extract_entries(
+                'http://list.youku.com/show/episode', show_id,
                 note='Downloading playlist data page %d' % (idx + 1),
-                transform_source=lambda s: js_to_json(strip_jsonp(s)))['html']
-            video_urls = re.findall(
-                r'<div[^>]+class="p-thumb"[^<]+<a[^>]+href="([^"]+)"',
-                playlist_data)
-            new_entries = [
-                self.url_result(urljoin(url, video_url), YoukuIE.ie_key())
-                for video_url in video_urls]
+                query={
+                    'id': page_config['showid'],
+                    'stage': reload_id,
+                })
             entries.extend(new_entries)
-            if len(new_entries) < self._PAGE_SIZE:
-                break
 
         desc = self._html_search_meta('description', webpage, fatal=False)
         playlist_title = desc.split(',')[0] if desc else None

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2017.07.15'
+__version__ = '2017.07.23'